chore: upgrade fickling to 0.1.6 (#30495)

Co-authored-by: Stephen Zhou <[email protected]>

.claude/skills/frontend-testing/SKILL.md

@@ -1,13 +1,13 @@
 ---
-name: Dify Frontend Testing
-description: Generate Jest + React Testing Library tests for Dify frontend components, hooks, and utilities. Triggers on testing, spec files, coverage, Jest, RTL, unit tests, integration tests, or write/review test requests.
+name: frontend-testing
+description: Generate Vitest + React Testing Library tests for Dify frontend components, hooks, and utilities. Triggers on testing, spec files, coverage, Vitest, RTL, unit tests, integration tests, or write/review test requests.
 ---
 
 # Dify Frontend Testing Skill
 
 This skill enables Claude to generate high-quality, comprehensive frontend tests for the Dify project following established conventions and best practices.
 
-> **⚠️ Authoritative Source**: This skill is derived from `web/testing/testing.md`. When in doubt, always refer to that document as the canonical specification.
+> **⚠️ Authoritative Source**: This skill is derived from `web/testing/testing.md`. Use Vitest mock/timer APIs (`vi.*`).
 
 ## When to Apply This Skill
 
@@ -15,7 +15,7 @@ Apply this skill when the user:
 
 - Asks to **write tests** for a component, hook, or utility
 - Asks to **review existing tests** for completeness
-- Mentions **Jest**, **React Testing Library**, **RTL**, or **spec files**
+- Mentions **Vitest**, **React Testing Library**, **RTL**, or **spec files**
 - Requests **test coverage** improvement
 - Uses `pnpm analyze-component` output as context
 - Mentions **testing**, **unit tests**, or **integration tests** for frontend code
@@ -33,9 +33,9 @@ Apply this skill when the user:
 
 | Tool | Version | Purpose |
 |------|---------|---------|
-| Jest | 29.7 | Test runner |
+| Vitest | 4.0.16 | Test runner |
 | React Testing Library | 16.0 | Component testing |
-| happy-dom | - | Test environment |
+| jsdom | - | Test environment |
 | nock | 14.0 | HTTP mocking |
 | TypeScript | 5.x | Type safety |
 
@@ -46,13 +46,13 @@ Apply this skill when the user:
 pnpm test
 
 # Watch mode
-pnpm test -- --watch
+pnpm test:watch
 
 # Run specific file
-pnpm test -- path/to/file.spec.tsx
+pnpm test path/to/file.spec.tsx
 
 # Generate coverage report
-pnpm test -- --coverage
+pnpm test:coverage
 
 # Analyze component complexity
 pnpm analyze-component <path>
@@ -77,9 +77,9 @@ import Component from './index'
 // import { ChildComponent } from './child-component'
 
 // ✅ Mock external dependencies only
-jest.mock('@/service/api')
-jest.mock('next/navigation', () => ({
-  useRouter: () => ({ push: jest.fn() }),
+vi.mock('@/service/api')
+vi.mock('next/navigation', () => ({
+  useRouter: () => ({ push: vi.fn() }),
   usePathname: () => '/test',
 }))
 
@@ -88,7 +88,7 @@ let mockSharedState = false
 
 describe('ComponentName', () => {
   beforeEach(() => {
-    jest.clearAllMocks()  // ✅ Reset mocks BEFORE each test
+    vi.clearAllMocks()  // ✅ Reset mocks BEFORE each test
     mockSharedState = false  // ✅ Reset shared state
   })
 
@@ -117,7 +117,7 @@ describe('ComponentName', () => {
   // User Interactions
   describe('User Interactions', () => {
     it('should handle click events', () => {
-      const handleClick = jest.fn()
+      const handleClick = vi.fn()
       render(<Component onClick={handleClick} />)
       
       fireEvent.click(screen.getByRole('button'))
@@ -155,7 +155,7 @@ describe('ComponentName', () => {
 For each file:
   ┌────────────────────────────────────────┐
   │ 1. Write test                          │
-  │ 2. Run: pnpm test -- <file>.spec.tsx   │
+  │ 2. Run: pnpm test <file>.spec.tsx      │
   │ 3. PASS? → Mark complete, next file    │
   │    FAIL? → Fix first, then continue    │
   └────────────────────────────────────────┘
@@ -178,7 +178,7 @@ Process in this order for multi-file testing:
 - **500+ lines**: Consider splitting before testing
 - **Many dependencies**: Extract logic into hooks first
 
-> 📖 See `guides/workflow.md` for complete workflow details and todo list format.
+> 📖 See `references/workflow.md` for complete workflow details and todo list format.
 
 ## Testing Strategy
 
@@ -289,17 +289,18 @@ For each test file generated, aim for:
 - ✅ **>95%** branch coverage
 - ✅ **>95%** line coverage
 
-> **Note**: For multi-file directories, process one file at a time with full coverage each. See `guides/workflow.md`.
+> **Note**: For multi-file directories, process one file at a time with full coverage each. See `references/workflow.md`.
 
 ## Detailed Guides
 
 For more detailed information, refer to:
 
-- `guides/workflow.md` - **Incremental testing workflow** (MUST READ for multi-file testing)
-- `guides/mocking.md` - Mock patterns and best practices
-- `guides/async-testing.md` - Async operations and API calls
-- `guides/domain-components.md` - Workflow, Dataset, Configuration testing
-- `guides/common-patterns.md` - Frequently used testing patterns
+- `references/workflow.md` - **Incremental testing workflow** (MUST READ for multi-file testing)
+- `references/mocking.md` - Mock patterns and best practices
+- `references/async-testing.md` - Async operations and API calls
+- `references/domain-components.md` - Workflow, Dataset, Configuration testing
+- `references/common-patterns.md` - Frequently used testing patterns
+- `references/checklist.md` - Test generation checklist and validation steps
 
 ## Authoritative References
 
@@ -315,7 +316,7 @@ For more detailed information, refer to:
 
 ### Project Configuration
 
-- `web/jest.config.ts` - Jest configuration
-- `web/jest.setup.ts` - Test environment setup
+- `web/vitest.config.ts` - Vitest configuration
+- `web/vitest.setup.ts` - Test environment setup
 - `web/testing/analyze-component.js` - Component analysis tool
-- `web/__mocks__/react-i18next.ts` - Shared i18n mock (auto-loaded by Jest, no explicit mock needed; override locally only for custom translations)
+- Modules are not mocked automatically. Global mocks live in `web/vitest.setup.ts` (for example `react-i18next`, `next/image`); mock other modules like `ky` or `mime` locally in test files.

.claude/skills/frontend-testing/assets/component-test.template.tsx

@@ -23,14 +23,14 @@ import userEvent from '@testing-library/user-event'
 // ============================================================================
 // Mocks
 // ============================================================================
-// WHY: Mocks must be hoisted to top of file (Jest requirement).
+// WHY: Mocks must be hoisted to top of file (Vitest requirement).
 // They run BEFORE imports, so keep them before component imports.
 
 // i18n (automatically mocked)
-// WHY: Shared mock at web/__mocks__/react-i18next.ts is auto-loaded by Jest
+// WHY: Global mock in web/vitest.setup.ts is auto-loaded by Vitest setup
 // No explicit mock needed - it returns translation keys as-is
 // Override only if custom translations are required:
-// jest.mock('react-i18next', () => ({
+// vi.mock('react-i18next', () => ({
 //   useTranslation: () => ({
 //     t: (key: string) => {
 //       const customTranslations: Record<string, string> = {
@@ -43,17 +43,17 @@ import userEvent from '@testing-library/user-event'
 
 // Router (if component uses useRouter, usePathname, useSearchParams)
 // WHY: Isolates tests from Next.js routing, enables testing navigation behavior
-// const mockPush = jest.fn()
-// jest.mock('next/navigation', () => ({
+// const mockPush = vi.fn()
+// vi.mock('next/navigation', () => ({
 //   useRouter: () => ({ push: mockPush }),
 //   usePathname: () => '/test-path',
 // }))
 
 // API services (if component fetches data)
 // WHY: Prevents real network calls, enables testing all states (loading/success/error)
-// jest.mock('@/service/api')
+// vi.mock('@/service/api')
 // import * as api from '@/service/api'
-// const mockedApi = api as jest.Mocked<typeof api>
+// const mockedApi = vi.mocked(api)
 
 // Shared mock state (for portal/dropdown components)
 // WHY: Portal components like PortalToFollowElem need shared state between
@@ -98,7 +98,7 @@ describe('ComponentName', () => {
   // - Prevents mock call history from leaking between tests
   // - MUST be beforeEach (not afterEach) to reset BEFORE assertions like toHaveBeenCalledTimes
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
     // Reset shared mock state if used (CRITICAL for portal/dropdown tests)
     // mockOpenState = false
   })
@@ -155,7 +155,7 @@ describe('ComponentName', () => {
       // - userEvent simulates real user behavior (focus, hover, then click)
       // - fireEvent is lower-level, doesn't trigger all browser events
       // const user = userEvent.setup()
-      // const handleClick = jest.fn()
+      // const handleClick = vi.fn()
       // render(<ComponentName onClick={handleClick} />)
       //
       // await user.click(screen.getByRole('button'))
@@ -165,7 +165,7 @@ describe('ComponentName', () => {
 
     it('should call onChange when value changes', async () => {
       // const user = userEvent.setup()
-      // const handleChange = jest.fn()
+      // const handleChange = vi.fn()
       // render(<ComponentName onChange={handleChange} />)
       //
       // await user.type(screen.getByRole('textbox'), 'new value')
@@ -198,7 +198,7 @@ describe('ComponentName', () => {
   })
 
   // --------------------------------------------------------------------------
-  // Async Operations (if component fetches data - useSWR, useQuery, fetch)
+  // Async Operations (if component fetches data - useQuery, fetch)
   // --------------------------------------------------------------------------
   // WHY: Async operations have 3 states users experience: loading, success, error
   describe('Async Operations', () => {

.claude/skills/frontend-testing/assets/hook-test.template.ts

@@ -15,9 +15,9 @@ import { renderHook, act, waitFor } from '@testing-library/react'
 // ============================================================================
 
 // API services (if hook fetches data)
-// jest.mock('@/service/api')
+// vi.mock('@/service/api')
 // import * as api from '@/service/api'
-// const mockedApi = api as jest.Mocked<typeof api>
+// const mockedApi = vi.mocked(api)
 
 // ============================================================================
 // Test Helpers
@@ -38,7 +38,7 @@ import { renderHook, act, waitFor } from '@testing-library/react'
 
 describe('useHookName', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
   })
 
   // --------------------------------------------------------------------------
@@ -145,7 +145,7 @@ describe('useHookName', () => {
   // --------------------------------------------------------------------------
   describe('Side Effects', () => {
     it('should call callback when value changes', () => {
-      // const callback = jest.fn()
+      // const callback = vi.fn()
       // const { result } = renderHook(() => useHookName({ onChange: callback }))
       //
       // act(() => {
@@ -156,9 +156,9 @@ describe('useHookName', () => {
     })
 
     it('should cleanup on unmount', () => {
-      // const cleanup = jest.fn()
-      // jest.spyOn(window, 'addEventListener')
-      // jest.spyOn(window, 'removeEventListener')
+      // const cleanup = vi.fn()
+      // vi.spyOn(window, 'addEventListener')
+      // vi.spyOn(window, 'removeEventListener')
       //
       // const { unmount } = renderHook(() => useHookName())
       //

.claude/skills/frontend-testing/references/async-testing.md

@@ -49,7 +49,7 @@ import userEvent from '@testing-library/user-event'
 
 it('should submit form', async () => {
   const user = userEvent.setup()
-  const onSubmit = jest.fn()
+  const onSubmit = vi.fn()
   
   render(<Form onSubmit={onSubmit} />)
   
@@ -77,15 +77,15 @@ it('should submit form', async () => {
 ```typescript
 describe('Debounced Search', () => {
   beforeEach(() => {
-    jest.useFakeTimers()
+    vi.useFakeTimers()
   })
 
   afterEach(() => {
-    jest.useRealTimers()
+    vi.useRealTimers()
   })
 
   it('should debounce search input', async () => {
-    const onSearch = jest.fn()
+    const onSearch = vi.fn()
     render(<SearchInput onSearch={onSearch} debounceMs={300} />)
     
     // Type in the input
@@ -95,7 +95,7 @@ describe('Debounced Search', () => {
     expect(onSearch).not.toHaveBeenCalled()
     
     // Advance timers
-    jest.advanceTimersByTime(300)
+    vi.advanceTimersByTime(300)
     
     // Now search is called
     expect(onSearch).toHaveBeenCalledWith('query')
@@ -107,8 +107,8 @@ describe('Debounced Search', () => {
 
 ```typescript
 it('should retry on failure', async () => {
-  jest.useFakeTimers()
-  const fetchData = jest.fn()
+  vi.useFakeTimers()
+  const fetchData = vi.fn()
     .mockRejectedValueOnce(new Error('Network error'))
     .mockResolvedValueOnce({ data: 'success' })
   
@@ -120,7 +120,7 @@ it('should retry on failure', async () => {
   })
   
   // Advance timer for retry
-  jest.advanceTimersByTime(1000)
+  vi.advanceTimersByTime(1000)
   
   // Second call succeeds
   await waitFor(() => {
@@ -128,7 +128,7 @@ it('should retry on failure', async () => {
     expect(screen.getByText('success')).toBeInTheDocument()
   })
   
-  jest.useRealTimers()
+  vi.useRealTimers()
 })
 ```
 
@@ -136,19 +136,19 @@ it('should retry on failure', async () => {
 
 ```typescript
 // Run all pending timers
-jest.runAllTimers()
+vi.runAllTimers()
 
 // Run only pending timers (not new ones created during execution)
-jest.runOnlyPendingTimers()
+vi.runOnlyPendingTimers()
 
 // Advance by specific time
-jest.advanceTimersByTime(1000)
+vi.advanceTimersByTime(1000)
 
 // Get current fake time
-jest.now()
+Date.now()
 
 // Clear all timers
-jest.clearAllTimers()
+vi.clearAllTimers()
 ```
 
 ## API Testing Patterns
@@ -158,7 +158,7 @@ jest.clearAllTimers()
 ```typescript
 describe('DataFetcher', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
   })
 
   it('should show loading state', () => {
@@ -241,7 +241,7 @@ it('should submit form and show success', async () => {
 
 ```typescript
 it('should fetch data on mount', async () => {
-  const fetchData = jest.fn().mockResolvedValue({ data: 'test' })
+  const fetchData = vi.fn().mockResolvedValue({ data: 'test' })
   
   render(<ComponentWithEffect fetchData={fetchData} />)
   
@@ -255,7 +255,7 @@ it('should fetch data on mount', async () => {
 
 ```typescript
 it('should refetch when id changes', async () => {
-  const fetchData = jest.fn().mockResolvedValue({ data: 'test' })
+  const fetchData = vi.fn().mockResolvedValue({ data: 'test' })
   
   const { rerender } = render(<ComponentWithEffect id="1" fetchData={fetchData} />)
   
@@ -276,8 +276,8 @@ it('should refetch when id changes', async () => {
 
 ```typescript
 it('should cleanup subscription on unmount', () => {
-  const subscribe = jest.fn()
-  const unsubscribe = jest.fn()
+  const subscribe = vi.fn()
+  const unsubscribe = vi.fn()
   subscribe.mockReturnValue(unsubscribe)
   
   const { unmount } = render(<SubscriptionComponent subscribe={subscribe} />)
@@ -332,14 +332,14 @@ expect(description).toBeInTheDocument()
 
 ```typescript
 // Bad - fake timers don't work well with real Promises
-jest.useFakeTimers()
+vi.useFakeTimers()
 await waitFor(() => {
   expect(screen.getByText('Data')).toBeInTheDocument()
 }) // May timeout!
 
 // Good - use runAllTimers or advanceTimersByTime
-jest.useFakeTimers()
+vi.useFakeTimers()
 render(<Component />)
-jest.runAllTimers()
+vi.runAllTimers()
 expect(screen.getByText('Data')).toBeInTheDocument()
 ```

.claude/skills/frontend-testing/references/checklist.md

@@ -74,9 +74,9 @@ Use this checklist when generating or reviewing tests for Dify frontend componen
 ### Mocks
 
 - [ ] **DO NOT mock base components** (`@/app/components/base/*`)
-- [ ] `jest.clearAllMocks()` in `beforeEach` (not `afterEach`)
+- [ ] `vi.clearAllMocks()` in `beforeEach` (not `afterEach`)
 - [ ] Shared mock state reset in `beforeEach`
-- [ ] i18n uses shared mock (auto-loaded); only override locally for custom translations
+- [ ] i18n uses global mock (auto-loaded in `web/vitest.setup.ts`); only override locally for custom translations
 - [ ] Router mocks match actual Next.js API
 - [ ] Mocks reflect actual component conditional behavior
 - [ ] Only mock: API services, complex context providers, third-party libs
@@ -114,15 +114,15 @@ For the current file being tested:
 
 **Run these checks after EACH test file, not just at the end:**
 
-- [ ] Run `pnpm test -- path/to/file.spec.tsx` - **MUST PASS before next file**
+- [ ] Run `pnpm test path/to/file.spec.tsx` - **MUST PASS before next file**
 - [ ] Fix any failures immediately
 - [ ] Mark file as complete in todo list
 - [ ] Only then proceed to next file
 
 ### After All Files Complete
 
-- [ ] Run full directory test: `pnpm test -- path/to/directory/`
-- [ ] Check coverage report: `pnpm test -- --coverage`
+- [ ] Run full directory test: `pnpm test path/to/directory/`
+- [ ] Check coverage report: `pnpm test:coverage`
 - [ ] Run `pnpm lint:fix` on all test files
 - [ ] Run `pnpm type-check:tsgo`
 
@@ -132,10 +132,10 @@ For the current file being tested:
 
 ```typescript
 // ❌ Mock doesn't match actual behavior
-jest.mock('./Component', () => () => <div>Mocked</div>)
+vi.mock('./Component', () => () => <div>Mocked</div>)
 
 // ✅ Mock matches actual conditional logic
-jest.mock('./Component', () => ({ isOpen }: any) =>
+vi.mock('./Component', () => ({ isOpen }: any) =>
   isOpen ? <div>Content</div> : null
 )
 ```
@@ -145,7 +145,7 @@ jest.mock('./Component', () => ({ isOpen }: any) =>
 ```typescript
 // ❌ Shared state not reset
 let mockState = false
-jest.mock('./useHook', () => () => mockState)
+vi.mock('./useHook', () => () => mockState)
 
 // ✅ Reset in beforeEach
 beforeEach(() => {
@@ -186,16 +186,16 @@ Always test these scenarios:
 
 ```bash
 # Run specific test
-pnpm test -- path/to/file.spec.tsx
+pnpm test path/to/file.spec.tsx
 
 # Run with coverage
-pnpm test -- --coverage path/to/file.spec.tsx
+pnpm test:coverage path/to/file.spec.tsx
 
 # Watch mode
-pnpm test -- --watch path/to/file.spec.tsx
+pnpm test:watch path/to/file.spec.tsx
 
 # Update snapshots (use sparingly)
-pnpm test -- -u path/to/file.spec.tsx
+pnpm test -u path/to/file.spec.tsx
 
 # Analyze component
 pnpm analyze-component path/to/component.tsx

.claude/skills/frontend-testing/references/common-patterns.md

@@ -126,7 +126,7 @@ describe('Counter', () => {
 describe('ControlledInput', () => {
   it('should call onChange with new value', async () => {
     const user = userEvent.setup()
-    const handleChange = jest.fn()
+    const handleChange = vi.fn()
     
     render(<ControlledInput value="" onChange={handleChange} />)
     
@@ -136,7 +136,7 @@ describe('ControlledInput', () => {
   })
 
   it('should display controlled value', () => {
-    render(<ControlledInput value="controlled" onChange={jest.fn()} />)
+    render(<ControlledInput value="controlled" onChange={vi.fn()} />)
     
     expect(screen.getByRole('textbox')).toHaveValue('controlled')
   })
@@ -195,7 +195,7 @@ describe('ItemList', () => {
 
   it('should handle item selection', async () => {
     const user = userEvent.setup()
-    const onSelect = jest.fn()
+    const onSelect = vi.fn()
     
     render(<ItemList items={items} onSelect={onSelect} />)
     
@@ -217,20 +217,20 @@ describe('ItemList', () => {
 ```typescript
 describe('Modal', () => {
   it('should not render when closed', () => {
-    render(<Modal isOpen={false} onClose={jest.fn()} />)
+    render(<Modal isOpen={false} onClose={vi.fn()} />)
     
     expect(screen.queryByRole('dialog')).not.toBeInTheDocument()
   })
 
   it('should render when open', () => {
-    render(<Modal isOpen={true} onClose={jest.fn()} />)
+    render(<Modal isOpen={true} onClose={vi.fn()} />)
     
     expect(screen.getByRole('dialog')).toBeInTheDocument()
   })
 
   it('should call onClose when clicking overlay', async () => {
     const user = userEvent.setup()
-    const handleClose = jest.fn()
+    const handleClose = vi.fn()
     
     render(<Modal isOpen={true} onClose={handleClose} />)
     
@@ -241,7 +241,7 @@ describe('Modal', () => {
 
   it('should call onClose when pressing Escape', async () => {
     const user = userEvent.setup()
-    const handleClose = jest.fn()
+    const handleClose = vi.fn()
     
     render(<Modal isOpen={true} onClose={handleClose} />)
     
@@ -254,7 +254,7 @@ describe('Modal', () => {
     const user = userEvent.setup()
     
     render(
-      <Modal isOpen={true} onClose={jest.fn()}>
+      <Modal isOpen={true} onClose={vi.fn()}>
         <button>First</button>
         <button>Second</button>
       </Modal>
@@ -279,7 +279,7 @@ describe('Modal', () => {
 describe('LoginForm', () => {
   it('should submit valid form', async () => {
     const user = userEvent.setup()
-    const onSubmit = jest.fn()
+    const onSubmit = vi.fn()
     
     render(<LoginForm onSubmit={onSubmit} />)
     
@@ -296,7 +296,7 @@ describe('LoginForm', () => {
   it('should show validation errors', async () => {
     const user = userEvent.setup()
     
-    render(<LoginForm onSubmit={jest.fn()} />)
+    render(<LoginForm onSubmit={vi.fn()} />)
     
     // Submit empty form
     await user.click(screen.getByRole('button', { name: /sign in/i }))
@@ -308,7 +308,7 @@ describe('LoginForm', () => {
   it('should validate email format', async () => {
     const user = userEvent.setup()
     
-    render(<LoginForm onSubmit={jest.fn()} />)
+    render(<LoginForm onSubmit={vi.fn()} />)
     
     await user.type(screen.getByLabelText(/email/i), 'invalid-email')
     await user.click(screen.getByRole('button', { name: /sign in/i }))
@@ -318,7 +318,7 @@ describe('LoginForm', () => {
 
   it('should disable submit button while submitting', async () => {
     const user = userEvent.setup()
-    const onSubmit = jest.fn(() => new Promise(resolve => setTimeout(resolve, 100)))
+    const onSubmit = vi.fn(() => new Promise(resolve => setTimeout(resolve, 100)))
     
     render(<LoginForm onSubmit={onSubmit} />)
     
@@ -407,7 +407,7 @@ it('test 1', () => {
 
 // Good - cleanup is automatic with RTL, but reset mocks
 beforeEach(() => {
-  jest.clearAllMocks()
+  vi.clearAllMocks()
 })
 ```
 

.claude/skills/frontend-testing/references/domain-components.md

@@ -23,7 +23,7 @@ import NodeConfigPanel from './node-config-panel'
 import { createMockNode, createMockWorkflowContext } from '@/__mocks__/workflow'
 
 // Mock workflow context
-jest.mock('@/app/components/workflow/hooks', () => ({
+vi.mock('@/app/components/workflow/hooks', () => ({
   useWorkflowStore: () => mockWorkflowStore,
   useNodesInteractions: () => mockNodesInteractions,
 }))
@@ -31,21 +31,21 @@ jest.mock('@/app/components/workflow/hooks', () => ({
 let mockWorkflowStore = {
   nodes: [],
   edges: [],
-  updateNode: jest.fn(),
+  updateNode: vi.fn(),
 }
 
 let mockNodesInteractions = {
-  handleNodeSelect: jest.fn(),
-  handleNodeDelete: jest.fn(),
+  handleNodeSelect: vi.fn(),
+  handleNodeDelete: vi.fn(),
 }
 
 describe('NodeConfigPanel', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
     mockWorkflowStore = {
       nodes: [],
       edges: [],
-      updateNode: jest.fn(),
+      updateNode: vi.fn(),
     }
   })
 
@@ -161,23 +161,23 @@ import { render, screen, fireEvent, waitFor } from '@testing-library/react'
 import userEvent from '@testing-library/user-event'
 import DocumentUploader from './document-uploader'
 
-jest.mock('@/service/datasets', () => ({
-  uploadDocument: jest.fn(),
-  parseDocument: jest.fn(),
+vi.mock('@/service/datasets', () => ({
+  uploadDocument: vi.fn(),
+  parseDocument: vi.fn(),
 }))
 
 import * as datasetService from '@/service/datasets'
-const mockedService = datasetService as jest.Mocked<typeof datasetService>
+const mockedService = vi.mocked(datasetService)
 
 describe('DocumentUploader', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
   })
 
   describe('File Upload', () => {
     it('should accept valid file types', async () => {
       const user = userEvent.setup()
-      const onUpload = jest.fn()
+      const onUpload = vi.fn()
       mockedService.uploadDocument.mockResolvedValue({ id: 'doc-1' })
       
       render(<DocumentUploader onUpload={onUpload} />)
@@ -326,14 +326,14 @@ describe('DocumentList', () => {
   describe('Search & Filtering', () => {
     it('should filter by search query', async () => {
       const user = userEvent.setup()
-      jest.useFakeTimers()
+      vi.useFakeTimers()
       
       render(<DocumentList datasetId="ds-1" />)
       
       await user.type(screen.getByPlaceholderText(/search/i), 'test query')
       
       // Debounce
-      jest.advanceTimersByTime(300)
+      vi.advanceTimersByTime(300)
       
       await waitFor(() => {
         expect(mockedService.getDocuments).toHaveBeenCalledWith(
@@ -342,7 +342,7 @@ describe('DocumentList', () => {
         )
       })
       
-      jest.useRealTimers()
+      vi.useRealTimers()
     })
   })
 })
@@ -367,13 +367,13 @@ import { render, screen, fireEvent, waitFor } from '@testing-library/react'
 import userEvent from '@testing-library/user-event'
 import AppConfigForm from './app-config-form'
 
-jest.mock('@/service/apps', () => ({
-  updateAppConfig: jest.fn(),
-  getAppConfig: jest.fn(),
+vi.mock('@/service/apps', () => ({
+  updateAppConfig: vi.fn(),
+  getAppConfig: vi.fn(),
 }))
 
 import * as appService from '@/service/apps'
-const mockedService = appService as jest.Mocked<typeof appService>
+const mockedService = vi.mocked(appService)
 
 describe('AppConfigForm', () => {
   const defaultConfig = {
@@ -384,7 +384,7 @@ describe('AppConfigForm', () => {
   }
 
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
     mockedService.getAppConfig.mockResolvedValue(defaultConfig)
   })
 

.claude/skills/frontend-testing/references/mocking.md

@@ -19,8 +19,8 @@
 
 ```typescript
 // ❌ WRONG: Don't mock base components
-jest.mock('@/app/components/base/loading', () => () => <div>Loading</div>)
-jest.mock('@/app/components/base/button', () => ({ children }: any) => <button>{children}</button>)
+vi.mock('@/app/components/base/loading', () => () => <div>Loading</div>)
+vi.mock('@/app/components/base/button', () => ({ children }: any) => <button>{children}</button>)
 
 // ✅ CORRECT: Import and use real base components
 import Loading from '@/app/components/base/loading'
@@ -41,20 +41,23 @@ Only mock these categories:
 
 | Location | Purpose |
 |----------|---------|
-| `web/__mocks__/` | Reusable mocks shared across multiple test files |
-| Test file | Test-specific mocks, inline with `jest.mock()` |
+| `web/vitest.setup.ts` | Global mocks shared by all tests (for example `react-i18next`, `next/image`) |
+| `web/__mocks__/` | Reusable mock factories shared across multiple test files |
+| Test file | Test-specific mocks, inline with `vi.mock()` |
+
+Modules are not mocked automatically. Use `vi.mock` in test files, or add global mocks in `web/vitest.setup.ts`.
 
 ## Essential Mocks
 
-### 1. i18n (Auto-loaded via Shared Mock)
+### 1. i18n (Auto-loaded via Global Mock)
 
-A shared mock is available at `web/__mocks__/react-i18next.ts` and is auto-loaded by Jest.
+A global mock is defined in `web/vitest.setup.ts` and is auto-loaded by Vitest setup.
 **No explicit mock needed** for most tests - it returns translation keys as-is.
 
 For tests requiring custom translations, override the mock:
 
 ```typescript
-jest.mock('react-i18next', () => ({
+vi.mock('react-i18next', () => ({
   useTranslation: () => ({
     t: (key: string) => {
       const translations: Record<string, string> = {
@@ -69,15 +72,15 @@ jest.mock('react-i18next', () => ({
 ### 2. Next.js Router
 
 ```typescript
-const mockPush = jest.fn()
-const mockReplace = jest.fn()
+const mockPush = vi.fn()
+const mockReplace = vi.fn()
 
-jest.mock('next/navigation', () => ({
+vi.mock('next/navigation', () => ({
   useRouter: () => ({
     push: mockPush,
     replace: mockReplace,
-    back: jest.fn(),
-    prefetch: jest.fn(),
+    back: vi.fn(),
+    prefetch: vi.fn(),
   }),
   usePathname: () => '/current-path',
   useSearchParams: () => new URLSearchParams('?key=value'),
@@ -85,7 +88,7 @@ jest.mock('next/navigation', () => ({
 
 describe('Component', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
   })
 
   it('should navigate on click', () => {
@@ -102,7 +105,7 @@ describe('Component', () => {
 // ⚠️ Important: Use shared state for components that depend on each other
 let mockPortalOpenState = false
 
-jest.mock('@/app/components/base/portal-to-follow-elem', () => ({
+vi.mock('@/app/components/base/portal-to-follow-elem', () => ({
   PortalToFollowElem: ({ children, open, ...props }: any) => {
     mockPortalOpenState = open || false  // Update shared state
     return <div data-testid="portal" data-open={open}>{children}</div>
@@ -119,7 +122,7 @@ jest.mock('@/app/components/base/portal-to-follow-elem', () => ({
 
 describe('Component', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
     mockPortalOpenState = false  // ✅ Reset shared state
   })
 })
@@ -130,13 +133,13 @@ describe('Component', () => {
 ```typescript
 import * as api from '@/service/api'
 
-jest.mock('@/service/api')
+vi.mock('@/service/api')
 
-const mockedApi = api as jest.Mocked<typeof api>
+const mockedApi = vi.mocked(api)
 
 describe('Component', () => {
   beforeEach(() => {
-    jest.clearAllMocks()
+    vi.clearAllMocks()
     
     // Setup default mock implementation
     mockedApi.fetchData.mockResolvedValue({ data: [] })
@@ -239,32 +242,9 @@ describe('Component with Context', () => {
 })
 ```
 
-### 7. SWR / React Query
+### 7. React Query
 
 ```typescript
-// SWR
-jest.mock('swr', () => ({
-  __esModule: true,
-  default: jest.fn(),
-}))
-
-import useSWR from 'swr'
-const mockedUseSWR = useSWR as jest.Mock
-
-describe('Component with SWR', () => {
-  it('should show loading state', () => {
-    mockedUseSWR.mockReturnValue({
-      data: undefined,
-      error: undefined,
-      isLoading: true,
-    })
-    
-    render(<Component />)
-    expect(screen.getByText(/loading/i)).toBeInTheDocument()
-  })
-})
-
-// React Query
 import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
 
 const createTestQueryClient = () => new QueryClient({

.claude/skills/frontend-testing/references/workflow.md

@@ -35,7 +35,7 @@ When testing a **single component, hook, or utility**:
 2. Run `pnpm analyze-component <path>` (if available)
 3. Check complexity score and features detected
 4. Write the test file
-5. Run test: `pnpm test -- <file>.spec.tsx`
+5. Run test: `pnpm test <file>.spec.tsx`
 6. Fix any failures
 7. Verify coverage meets goals (100% function, >95% branch)
 ```
@@ -80,7 +80,7 @@ Process files in this recommended order:
 ```
 ┌─────────────────────────────────────────────┐
 │  1. Write test file                         │
-│  2. Run: pnpm test -- <file>.spec.tsx       │
+│  2. Run: pnpm test <file>.spec.tsx          │
 │  3. If FAIL → Fix immediately, re-run       │
 │  4. If PASS → Mark complete in todo list    │
 │  5. ONLY THEN proceed to next file          │
@@ -95,10 +95,10 @@ After all individual tests pass:
 
 ```bash
 # Run all tests in the directory together
-pnpm test -- path/to/directory/
+pnpm test path/to/directory/
 
 # Check coverage
-pnpm test -- --coverage path/to/directory/
+pnpm test:coverage path/to/directory/
 ```
 
 ## Component Complexity Guidelines
@@ -201,9 +201,9 @@ Run pnpm test  ← Multiple failures, hard to debug
 ```
 # GOOD: Incremental with verification
 Write component-a.spec.tsx
-Run pnpm test -- component-a.spec.tsx ✅
+Run pnpm test component-a.spec.tsx ✅
 Write component-b.spec.tsx
-Run pnpm test -- component-b.spec.tsx ✅
+Run pnpm test component-b.spec.tsx ✅
 ...continue...
 ```
 

.codex/skills

@@ -0,0 +1 @@
+../.claude/skills

.devcontainer/devcontainer.json

@@ -6,6 +6,9 @@
 		"context": "..",
 		"dockerfile": "Dockerfile"
 	},
+	"mounts": [
+		"source=dify-dev-tmp,target=/tmp,type=volume"
+	],
 	"features": {
 		"ghcr.io/devcontainers/features/node:1": {
 			"nodeGypDependencies": true,
@@ -34,19 +37,13 @@
 	},
 	"postStartCommand": "./.devcontainer/post_start_command.sh",
 	"postCreateCommand": "./.devcontainer/post_create_command.sh"
-
 	// Features to add to the dev container. More info: https://containers.dev/features.
 	// "features": {},
-
 	// Use 'forwardPorts' to make a list of ports inside the container available locally.
 	// "forwardPorts": [],
-
 	// Use 'postCreateCommand' to run commands after the container is created.
 	// "postCreateCommand": "python --version",
-
 	// Configure tool-specific properties.
 	// "customizations": {},
-
 	// Uncomment to connect as root instead. More info: https://aka.ms/dev-containers-non-root.
-	// "remoteUser": "root"
-}
+}

.devcontainer/post_create_command.sh

@@ -1,6 +1,7 @@
 #!/bin/bash
 WORKSPACE_ROOT=$(pwd)
 
+export COREPACK_ENABLE_DOWNLOAD_PROMPT=0
 corepack enable
 cd web && pnpm install
 pipx install uv

.github/CODEOWNERS

@@ -7,234 +7,243 @@
 * @crazywoola @laipz8200 @Yeuoly
 
 # CODEOWNERS file
-.github/CODEOWNERS @laipz8200 @crazywoola
+/.github/CODEOWNERS @laipz8200 @crazywoola
 
 # Docs
-docs/ @crazywoola
+/docs/ @crazywoola
 
 # Backend (default owner, more specific rules below will override)
-api/ @QuantumGhost
+/api/ @QuantumGhost
 
 # Backend - MCP
-api/core/mcp/ @Nov1c444
-api/core/entities/mcp_provider.py @Nov1c444
-api/services/tools/mcp_tools_manage_service.py @Nov1c444
-api/controllers/mcp/ @Nov1c444
-api/controllers/console/app/mcp_server.py @Nov1c444
-api/tests/**/*mcp* @Nov1c444
+/api/core/mcp/ @Nov1c444
+/api/core/entities/mcp_provider.py @Nov1c444
+/api/services/tools/mcp_tools_manage_service.py @Nov1c444
+/api/controllers/mcp/ @Nov1c444
+/api/controllers/console/app/mcp_server.py @Nov1c444
+/api/tests/**/*mcp* @Nov1c444
 
 # Backend - Workflow - Engine (Core graph execution engine)
-api/core/workflow/graph_engine/ @laipz8200 @QuantumGhost
-api/core/workflow/runtime/ @laipz8200 @QuantumGhost
-api/core/workflow/graph/ @laipz8200 @QuantumGhost
-api/core/workflow/graph_events/ @laipz8200 @QuantumGhost
-api/core/workflow/node_events/ @laipz8200 @QuantumGhost
-api/core/model_runtime/ @laipz8200 @QuantumGhost
+/api/core/workflow/graph_engine/ @laipz8200 @QuantumGhost
+/api/core/workflow/runtime/ @laipz8200 @QuantumGhost
+/api/core/workflow/graph/ @laipz8200 @QuantumGhost
+/api/core/workflow/graph_events/ @laipz8200 @QuantumGhost
+/api/core/workflow/node_events/ @laipz8200 @QuantumGhost
+/api/core/model_runtime/ @laipz8200 @QuantumGhost
 
 # Backend - Workflow - Nodes (Agent, Iteration, Loop, LLM)
-api/core/workflow/nodes/agent/ @Nov1c444
-api/core/workflow/nodes/iteration/ @Nov1c444
-api/core/workflow/nodes/loop/ @Nov1c444
-api/core/workflow/nodes/llm/ @Nov1c444
+/api/core/workflow/nodes/agent/ @Nov1c444
+/api/core/workflow/nodes/iteration/ @Nov1c444
+/api/core/workflow/nodes/loop/ @Nov1c444
+/api/core/workflow/nodes/llm/ @Nov1c444
 
 # Backend - RAG (Retrieval Augmented Generation)
-api/core/rag/ @JohnJyong
-api/services/rag_pipeline/ @JohnJyong
-api/services/dataset_service.py @JohnJyong
-api/services/knowledge_service.py @JohnJyong
-api/services/external_knowledge_service.py @JohnJyong
-api/services/hit_testing_service.py @JohnJyong
-api/services/metadata_service.py @JohnJyong
-api/services/vector_service.py @JohnJyong
-api/services/entities/knowledge_entities/ @JohnJyong
-api/services/entities/external_knowledge_entities/ @JohnJyong
-api/controllers/console/datasets/ @JohnJyong
-api/controllers/service_api/dataset/ @JohnJyong
-api/models/dataset.py @JohnJyong
-api/tasks/rag_pipeline/ @JohnJyong
-api/tasks/add_document_to_index_task.py @JohnJyong
-api/tasks/batch_clean_document_task.py @JohnJyong
-api/tasks/clean_document_task.py @JohnJyong
-api/tasks/clean_notion_document_task.py @JohnJyong
-api/tasks/document_indexing_task.py @JohnJyong
-api/tasks/document_indexing_sync_task.py @JohnJyong
-api/tasks/document_indexing_update_task.py @JohnJyong
-api/tasks/duplicate_document_indexing_task.py @JohnJyong
-api/tasks/recover_document_indexing_task.py @JohnJyong
-api/tasks/remove_document_from_index_task.py @JohnJyong
-api/tasks/retry_document_indexing_task.py @JohnJyong
-api/tasks/sync_website_document_indexing_task.py @JohnJyong
-api/tasks/batch_create_segment_to_index_task.py @JohnJyong
-api/tasks/create_segment_to_index_task.py @JohnJyong
-api/tasks/delete_segment_from_index_task.py @JohnJyong
-api/tasks/disable_segment_from_index_task.py @JohnJyong
-api/tasks/disable_segments_from_index_task.py @JohnJyong
-api/tasks/enable_segment_to_index_task.py @JohnJyong
-api/tasks/enable_segments_to_index_task.py @JohnJyong
-api/tasks/clean_dataset_task.py @JohnJyong
-api/tasks/deal_dataset_index_update_task.py @JohnJyong
-api/tasks/deal_dataset_vector_index_task.py @JohnJyong
+/api/core/rag/ @JohnJyong
+/api/services/rag_pipeline/ @JohnJyong
+/api/services/dataset_service.py @JohnJyong
+/api/services/knowledge_service.py @JohnJyong
+/api/services/external_knowledge_service.py @JohnJyong
+/api/services/hit_testing_service.py @JohnJyong
+/api/services/metadata_service.py @JohnJyong
+/api/services/vector_service.py @JohnJyong
+/api/services/entities/knowledge_entities/ @JohnJyong
+/api/services/entities/external_knowledge_entities/ @JohnJyong
+/api/controllers/console/datasets/ @JohnJyong
+/api/controllers/service_api/dataset/ @JohnJyong
+/api/models/dataset.py @JohnJyong
+/api/tasks/rag_pipeline/ @JohnJyong
+/api/tasks/add_document_to_index_task.py @JohnJyong
+/api/tasks/batch_clean_document_task.py @JohnJyong
+/api/tasks/clean_document_task.py @JohnJyong
+/api/tasks/clean_notion_document_task.py @JohnJyong
+/api/tasks/document_indexing_task.py @JohnJyong
+/api/tasks/document_indexing_sync_task.py @JohnJyong
+/api/tasks/document_indexing_update_task.py @JohnJyong
+/api/tasks/duplicate_document_indexing_task.py @JohnJyong
+/api/tasks/recover_document_indexing_task.py @JohnJyong
+/api/tasks/remove_document_from_index_task.py @JohnJyong
+/api/tasks/retry_document_indexing_task.py @JohnJyong
+/api/tasks/sync_website_document_indexing_task.py @JohnJyong
+/api/tasks/batch_create_segment_to_index_task.py @JohnJyong
+/api/tasks/create_segment_to_index_task.py @JohnJyong
+/api/tasks/delete_segment_from_index_task.py @JohnJyong
+/api/tasks/disable_segment_from_index_task.py @JohnJyong
+/api/tasks/disable_segments_from_index_task.py @JohnJyong
+/api/tasks/enable_segment_to_index_task.py @JohnJyong
+/api/tasks/enable_segments_to_index_task.py @JohnJyong
+/api/tasks/clean_dataset_task.py @JohnJyong
+/api/tasks/deal_dataset_index_update_task.py @JohnJyong
+/api/tasks/deal_dataset_vector_index_task.py @JohnJyong
 
 # Backend - Plugins
-api/core/plugin/ @Mairuis @Yeuoly @Stream29
-api/services/plugin/ @Mairuis @Yeuoly @Stream29
-api/controllers/console/workspace/plugin.py @Mairuis @Yeuoly @Stream29
-api/controllers/inner_api/plugin/ @Mairuis @Yeuoly @Stream29
-api/tasks/process_tenant_plugin_autoupgrade_check_task.py @Mairuis @Yeuoly @Stream29
+/api/core/plugin/ @Mairuis @Yeuoly @Stream29
+/api/services/plugin/ @Mairuis @Yeuoly @Stream29
+/api/controllers/console/workspace/plugin.py @Mairuis @Yeuoly @Stream29
+/api/controllers/inner_api/plugin/ @Mairuis @Yeuoly @Stream29
+/api/tasks/process_tenant_plugin_autoupgrade_check_task.py @Mairuis @Yeuoly @Stream29
 
 # Backend - Trigger/Schedule/Webhook
-api/controllers/trigger/ @Mairuis @Yeuoly
-api/controllers/console/app/workflow_trigger.py @Mairuis @Yeuoly
-api/controllers/console/workspace/trigger_providers.py @Mairuis @Yeuoly
-api/core/trigger/ @Mairuis @Yeuoly
-api/core/app/layers/trigger_post_layer.py @Mairuis @Yeuoly
-api/services/trigger/ @Mairuis @Yeuoly
-api/models/trigger.py @Mairuis @Yeuoly
-api/fields/workflow_trigger_fields.py @Mairuis @Yeuoly
-api/repositories/workflow_trigger_log_repository.py @Mairuis @Yeuoly
-api/repositories/sqlalchemy_workflow_trigger_log_repository.py @Mairuis @Yeuoly
-api/libs/schedule_utils.py @Mairuis @Yeuoly
-api/services/workflow/scheduler.py @Mairuis @Yeuoly
-api/schedule/trigger_provider_refresh_task.py @Mairuis @Yeuoly
-api/schedule/workflow_schedule_task.py @Mairuis @Yeuoly
-api/tasks/trigger_processing_tasks.py @Mairuis @Yeuoly
-api/tasks/trigger_subscription_refresh_tasks.py @Mairuis @Yeuoly
-api/tasks/workflow_schedule_tasks.py @Mairuis @Yeuoly
-api/tasks/workflow_cfs_scheduler/ @Mairuis @Yeuoly
-api/events/event_handlers/sync_plugin_trigger_when_app_created.py @Mairuis @Yeuoly
-api/events/event_handlers/update_app_triggers_when_app_published_workflow_updated.py @Mairuis @Yeuoly
-api/events/event_handlers/sync_workflow_schedule_when_app_published.py @Mairuis @Yeuoly
-api/events/event_handlers/sync_webhook_when_app_created.py @Mairuis @Yeuoly
+/api/controllers/trigger/ @Mairuis @Yeuoly
+/api/controllers/console/app/workflow_trigger.py @Mairuis @Yeuoly
+/api/controllers/console/workspace/trigger_providers.py @Mairuis @Yeuoly
+/api/core/trigger/ @Mairuis @Yeuoly
+/api/core/app/layers/trigger_post_layer.py @Mairuis @Yeuoly
+/api/services/trigger/ @Mairuis @Yeuoly
+/api/models/trigger.py @Mairuis @Yeuoly
+/api/fields/workflow_trigger_fields.py @Mairuis @Yeuoly
+/api/repositories/workflow_trigger_log_repository.py @Mairuis @Yeuoly
+/api/repositories/sqlalchemy_workflow_trigger_log_repository.py @Mairuis @Yeuoly
+/api/libs/schedule_utils.py @Mairuis @Yeuoly
+/api/services/workflow/scheduler.py @Mairuis @Yeuoly
+/api/schedule/trigger_provider_refresh_task.py @Mairuis @Yeuoly
+/api/schedule/workflow_schedule_task.py @Mairuis @Yeuoly
+/api/tasks/trigger_processing_tasks.py @Mairuis @Yeuoly
+/api/tasks/trigger_subscription_refresh_tasks.py @Mairuis @Yeuoly
+/api/tasks/workflow_schedule_tasks.py @Mairuis @Yeuoly
+/api/tasks/workflow_cfs_scheduler/ @Mairuis @Yeuoly
+/api/events/event_handlers/sync_plugin_trigger_when_app_created.py @Mairuis @Yeuoly
+/api/events/event_handlers/update_app_triggers_when_app_published_workflow_updated.py @Mairuis @Yeuoly
+/api/events/event_handlers/sync_workflow_schedule_when_app_published.py @Mairuis @Yeuoly
+/api/events/event_handlers/sync_webhook_when_app_created.py @Mairuis @Yeuoly
 
 # Backend - Async Workflow
-api/services/async_workflow_service.py @Mairuis @Yeuoly
-api/tasks/async_workflow_tasks.py @Mairuis @Yeuoly
+/api/services/async_workflow_service.py @Mairuis @Yeuoly
+/api/tasks/async_workflow_tasks.py @Mairuis @Yeuoly
 
 # Backend - Billing
-api/services/billing_service.py @hj24 @zyssyz123
-api/controllers/console/billing/ @hj24 @zyssyz123
+/api/services/billing_service.py @hj24 @zyssyz123
+/api/controllers/console/billing/ @hj24 @zyssyz123
 
 # Backend - Enterprise
-api/configs/enterprise/ @GarfieldDai @GareArc
-api/services/enterprise/ @GarfieldDai @GareArc
-api/services/feature_service.py @GarfieldDai @GareArc
-api/controllers/console/feature.py @GarfieldDai @GareArc
-api/controllers/web/feature.py @GarfieldDai @GareArc
+/api/configs/enterprise/ @GarfieldDai @GareArc
+/api/services/enterprise/ @GarfieldDai @GareArc
+/api/services/feature_service.py @GarfieldDai @GareArc
+/api/controllers/console/feature.py @GarfieldDai @GareArc
+/api/controllers/web/feature.py @GarfieldDai @GareArc
 
 # Backend - Database Migrations
-api/migrations/ @snakevash @laipz8200 @MRZHUH
+/api/migrations/ @snakevash @laipz8200 @MRZHUH
+
+# Backend - Vector DB Middleware
+/api/configs/middleware/vdb/* @JohnJyong
 
 # Frontend
-web/ @iamjoel
+/web/ @iamjoel
+
+# Frontend - Web Tests
+/.github/workflows/web-tests.yml @iamjoel
 
 # Frontend - App - Orchestration
-web/app/components/workflow/ @iamjoel @zxhlyh
-web/app/components/workflow-app/ @iamjoel @zxhlyh
-web/app/components/app/configuration/ @iamjoel @zxhlyh
-web/app/components/app/app-publisher/ @iamjoel @zxhlyh
+/web/app/components/workflow/ @iamjoel @zxhlyh
+/web/app/components/workflow-app/ @iamjoel @zxhlyh
+/web/app/components/app/configuration/ @iamjoel @zxhlyh
+/web/app/components/app/app-publisher/ @iamjoel @zxhlyh
 
 # Frontend - WebApp - Chat
-web/app/components/base/chat/ @iamjoel @zxhlyh
+/web/app/components/base/chat/ @iamjoel @zxhlyh
 
 # Frontend - WebApp - Completion
-web/app/components/share/text-generation/ @iamjoel @zxhlyh
+/web/app/components/share/text-generation/ @iamjoel @zxhlyh
 
 # Frontend - App - List and Creation
-web/app/components/apps/ @JzoNgKVO @iamjoel
-web/app/components/app/create-app-dialog/ @JzoNgKVO @iamjoel
-web/app/components/app/create-app-modal/ @JzoNgKVO @iamjoel
-web/app/components/app/create-from-dsl-modal/ @JzoNgKVO @iamjoel
+/web/app/components/apps/ @JzoNgKVO @iamjoel
+/web/app/components/app/create-app-dialog/ @JzoNgKVO @iamjoel
+/web/app/components/app/create-app-modal/ @JzoNgKVO @iamjoel
+/web/app/components/app/create-from-dsl-modal/ @JzoNgKVO @iamjoel
 
 # Frontend - App - API Documentation
-web/app/components/develop/ @JzoNgKVO @iamjoel
+/web/app/components/develop/ @JzoNgKVO @iamjoel
 
 # Frontend - App - Logs and Annotations
-web/app/components/app/workflow-log/ @JzoNgKVO @iamjoel
-web/app/components/app/log/ @JzoNgKVO @iamjoel
-web/app/components/app/log-annotation/ @JzoNgKVO @iamjoel
-web/app/components/app/annotation/ @JzoNgKVO @iamjoel
+/web/app/components/app/workflow-log/ @JzoNgKVO @iamjoel
+/web/app/components/app/log/ @JzoNgKVO @iamjoel
+/web/app/components/app/log-annotation/ @JzoNgKVO @iamjoel
+/web/app/components/app/annotation/ @JzoNgKVO @iamjoel
 
 # Frontend - App - Monitoring
-web/app/(commonLayout)/app/(appDetailLayout)/\[appId\]/overview/ @JzoNgKVO @iamjoel
-web/app/components/app/overview/ @JzoNgKVO @iamjoel
+/web/app/(commonLayout)/app/(appDetailLayout)/\[appId\]/overview/ @JzoNgKVO @iamjoel
+/web/app/components/app/overview/ @JzoNgKVO @iamjoel
 
 # Frontend - App - Settings
-web/app/components/app-sidebar/ @JzoNgKVO @iamjoel
+/web/app/components/app-sidebar/ @JzoNgKVO @iamjoel
 
 # Frontend - RAG - Hit Testing
-web/app/components/datasets/hit-testing/ @JzoNgKVO @iamjoel
+/web/app/components/datasets/hit-testing/ @JzoNgKVO @iamjoel
 
 # Frontend - RAG - List and Creation
-web/app/components/datasets/list/ @iamjoel @WTW0313
-web/app/components/datasets/create/ @iamjoel @WTW0313
-web/app/components/datasets/create-from-pipeline/ @iamjoel @WTW0313
-web/app/components/datasets/external-knowledge-base/ @iamjoel @WTW0313
+/web/app/components/datasets/list/ @iamjoel @WTW0313
+/web/app/components/datasets/create/ @iamjoel @WTW0313
+/web/app/components/datasets/create-from-pipeline/ @iamjoel @WTW0313
+/web/app/components/datasets/external-knowledge-base/ @iamjoel @WTW0313
 
 # Frontend - RAG - Orchestration (general rule first, specific rules below override)
-web/app/components/rag-pipeline/ @iamjoel @WTW0313
-web/app/components/rag-pipeline/components/rag-pipeline-main.tsx @iamjoel @zxhlyh
-web/app/components/rag-pipeline/store/ @iamjoel @zxhlyh
+/web/app/components/rag-pipeline/ @iamjoel @WTW0313
+/web/app/components/rag-pipeline/components/rag-pipeline-main.tsx @iamjoel @zxhlyh
+/web/app/components/rag-pipeline/store/ @iamjoel @zxhlyh
 
 # Frontend - RAG - Documents List
-web/app/components/datasets/documents/list.tsx @iamjoel @WTW0313
-web/app/components/datasets/documents/create-from-pipeline/ @iamjoel @WTW0313
+/web/app/components/datasets/documents/list.tsx @iamjoel @WTW0313
+/web/app/components/datasets/documents/create-from-pipeline/ @iamjoel @WTW0313
 
 # Frontend - RAG - Segments List
-web/app/components/datasets/documents/detail/ @iamjoel @WTW0313
+/web/app/components/datasets/documents/detail/ @iamjoel @WTW0313
 
 # Frontend - RAG - Settings
-web/app/components/datasets/settings/ @iamjoel @WTW0313
+/web/app/components/datasets/settings/ @iamjoel @WTW0313
 
 # Frontend - Ecosystem - Plugins
-web/app/components/plugins/ @iamjoel @zhsama
+/web/app/components/plugins/ @iamjoel @zhsama
 
 # Frontend - Ecosystem - Tools
-web/app/components/tools/ @iamjoel @Yessenia-d
+/web/app/components/tools/ @iamjoel @Yessenia-d
 
 # Frontend - Ecosystem - MarketPlace
-web/app/components/plugins/marketplace/ @iamjoel @Yessenia-d
+/web/app/components/plugins/marketplace/ @iamjoel @Yessenia-d
 
 # Frontend - Login and Registration
-web/app/signin/ @douxc @iamjoel
-web/app/signup/ @douxc @iamjoel
-web/app/reset-password/ @douxc @iamjoel
-web/app/install/ @douxc @iamjoel
-web/app/init/ @douxc @iamjoel
-web/app/forgot-password/ @douxc @iamjoel
-web/app/account/ @douxc @iamjoel
+/web/app/signin/ @douxc @iamjoel
+/web/app/signup/ @douxc @iamjoel
+/web/app/reset-password/ @douxc @iamjoel
+/web/app/install/ @douxc @iamjoel
+/web/app/init/ @douxc @iamjoel
+/web/app/forgot-password/ @douxc @iamjoel
+/web/app/account/ @douxc @iamjoel
 
 # Frontend - Service Authentication
-web/service/base.ts @douxc @iamjoel
+/web/service/base.ts @douxc @iamjoel
 
 # Frontend - WebApp Authentication and Access Control
-web/app/(shareLayout)/components/ @douxc @iamjoel
-web/app/(shareLayout)/webapp-signin/ @douxc @iamjoel
-web/app/(shareLayout)/webapp-reset-password/ @douxc @iamjoel
-web/app/components/app/app-access-control/ @douxc @iamjoel
+/web/app/(shareLayout)/components/ @douxc @iamjoel
+/web/app/(shareLayout)/webapp-signin/ @douxc @iamjoel
+/web/app/(shareLayout)/webapp-reset-password/ @douxc @iamjoel
+/web/app/components/app/app-access-control/ @douxc @iamjoel
 
 # Frontend - Explore Page
-web/app/components/explore/ @CodingOnStar @iamjoel
+/web/app/components/explore/ @CodingOnStar @iamjoel
 
 # Frontend - Personal Settings
-web/app/components/header/account-setting/ @CodingOnStar @iamjoel
-web/app/components/header/account-dropdown/ @CodingOnStar @iamjoel
+/web/app/components/header/account-setting/ @CodingOnStar @iamjoel
+/web/app/components/header/account-dropdown/ @CodingOnStar @iamjoel
 
 # Frontend - Analytics
-web/app/components/base/ga/ @CodingOnStar @iamjoel
+/web/app/components/base/ga/ @CodingOnStar @iamjoel
 
 # Frontend - Base Components
-web/app/components/base/ @iamjoel @zxhlyh
+/web/app/components/base/ @iamjoel @zxhlyh
 
 # Frontend - Utils and Hooks
-web/utils/classnames.ts @iamjoel @zxhlyh
-web/utils/time.ts @iamjoel @zxhlyh
-web/utils/format.ts @iamjoel @zxhlyh
-web/utils/clipboard.ts @iamjoel @zxhlyh
-web/hooks/use-document-title.ts @iamjoel @zxhlyh
+/web/utils/classnames.ts @iamjoel @zxhlyh
+/web/utils/time.ts @iamjoel @zxhlyh
+/web/utils/format.ts @iamjoel @zxhlyh
+/web/utils/clipboard.ts @iamjoel @zxhlyh
+/web/hooks/use-document-title.ts @iamjoel @zxhlyh
 
 # Frontend - Billing and Education
-web/app/components/billing/ @iamjoel @zxhlyh
-web/app/education-apply/ @iamjoel @zxhlyh
+/web/app/components/billing/ @iamjoel @zxhlyh
+/web/app/education-apply/ @iamjoel @zxhlyh
 
 # Frontend - Workspace
-web/app/components/header/account-dropdown/workplace-selector/ @iamjoel @zxhlyh
+/web/app/components/header/account-dropdown/workplace-selector/ @iamjoel @zxhlyh
+
+# Docker
+/docker/* @laipz8200

.github/workflows/autofix.yml

@@ -13,12 +13,28 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
+
+      - name: Check Docker Compose inputs
+        id: docker-compose-changes
+        uses: tj-actions/changed-files@v46
+        with:
+          files: |
+            docker/generate_docker_compose
+            docker/.env.example
+            docker/docker-compose-template.yaml
+            docker/docker-compose.yaml
       - uses: actions/setup-python@v5
         with:
           python-version: "3.11"
 
       - uses: astral-sh/setup-uv@v6
 
+      - name: Generate Docker Compose
+        if: steps.docker-compose-changes.outputs.any_changed == 'true'
+        run: |
+          cd docker
+          ./generate_docker_compose
+
       - run: |
           cd api
           uv sync --dev
@@ -66,27 +82,6 @@ jobs:
       # mdformat breaks YAML front matter in markdown files. Add --exclude for directories containing YAML front matter.
       - name: mdformat
         run: |
-          uvx --python 3.13 mdformat . --exclude ".claude/skills/**"
-
-      - name: Install pnpm
-        uses: pnpm/action-setup@v4
-        with:
-          package_json_file: web/package.json
-          run_install: false
-
-      - name: Setup NodeJS
-        uses: actions/setup-node@v4
-        with:
-          node-version: 22
-          cache: pnpm
-          cache-dependency-path: ./web/pnpm-lock.yaml
-
-      - name: Web dependencies
-        working-directory: ./web
-        run: pnpm install --frozen-lockfile
-
-      - name: oxlint
-        working-directory: ./web
-        run: pnpm exec oxlint --config .oxlintrc.json --fix .
+          uvx --python 3.13 mdformat . --exclude ".claude/skills/**/SKILL.md"
 
       - uses: autofix-ci/action@635ffb0c9798bd160680f18fd73371e355b85f27

.github/workflows/style.yml

@@ -88,7 +88,7 @@ jobs:
         uses: actions/setup-node@v4
         if: steps.changed-files.outputs.any_changed == 'true'
         with:
-          node-version: 22
+          node-version: 24
           cache: pnpm
           cache-dependency-path: ./web/pnpm-lock.yaml
 
@@ -108,36 +108,6 @@ jobs:
         working-directory: ./web
         run: pnpm run type-check:tsgo
 
-  docker-compose-template:
-    name: Docker Compose Template
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          persist-credentials: false
-
-      - name: Check changed files
-        id: changed-files
-        uses: tj-actions/changed-files@v46
-        with:
-          files: |
-            docker/generate_docker_compose
-            docker/.env.example
-            docker/docker-compose-template.yaml
-            docker/docker-compose.yaml
-
-      - name: Generate Docker Compose
-        if: steps.changed-files.outputs.any_changed == 'true'
-        run: |
-          cd docker
-          ./generate_docker_compose
-
-      - name: Check for changes
-        if: steps.changed-files.outputs.any_changed == 'true'
-        run: git diff --exit-code
-
   superlinter:
     name: SuperLinter
     runs-on: ubuntu-latest

.github/workflows/tool-test-sdks.yaml

@@ -16,10 +16,6 @@ jobs:
     name: unit test for Node.js SDK
     runs-on: ubuntu-latest
 
-    strategy:
-      matrix:
-        node-version: [16, 18, 20, 22]
-
     defaults:
       run:
         working-directory: sdks/nodejs-client
@@ -29,10 +25,15 @@ jobs:
         with:
           persist-credentials: false
 
+<<<<<<< HEAD
       - name: Use Node.js ${{ matrix.node-version }}
         uses: actions/setup-node@v4
+=======
+      - name: Use Node.js
+        uses: actions/setup-node@v6
+>>>>>>> 328897f81c (build: require node 24.13.0 (#30945))
         with:
-          node-version: ${{ matrix.node-version }}
+          node-version: 24
           cache: ''
           cache-dependency-path: 'pnpm-lock.yaml'
 

.github/workflows/translate-i18n-base-on-english.yml

@@ -1,4 +1,4 @@
-name: Check i18n Files and Create PR
+name: Translate i18n Files Based on English
 
 on:
   push:
@@ -67,25 +67,19 @@ jobs:
         working-directory: ./web
         run: pnpm run auto-gen-i18n ${{ env.FILE_ARGS }}
 
-      - name: Generate i18n type definitions
-        if: env.FILES_CHANGED == 'true'
-        working-directory: ./web
-        run: pnpm run gen:i18n-types
-
       - name: Create Pull Request
         if: env.FILES_CHANGED == 'true'
         uses: peter-evans/create-pull-request@v6
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           commit-message: 'chore(i18n): update translations based on en-US changes'
-          title: 'chore(i18n): translate i18n files and update type definitions'
+          title: 'chore(i18n): translate i18n files based on en-US changes'
           body: |
-            This PR was automatically created to update i18n files and TypeScript type definitions based on changes in en-US locale.
+            This PR was automatically created to update i18n translation files based on changes in en-US locale.
 
             **Triggered by:** ${{ github.sha }}
 
             **Changes included:**
             - Updated translation files for all locales
-            - Regenerated TypeScript type definitions for type safety
           branch: chore/automated-i18n-updates-${{ github.sha }}
           delete-branch: true

.github/workflows/translate-i18n-claude.yml

@@ -0,0 +1,421 @@
+name: Translate i18n Files with Claude Code
+
+# Note: claude-code-action doesn't support push events directly.
+# Push events are handled by trigger-i18n-sync.yml which sends repository_dispatch.
+# See: https://github.com/langgenius/dify/issues/30743
+
+on:
+  repository_dispatch:
+    types: [i18n-sync]
+  workflow_dispatch:
+    inputs:
+      files:
+        description: 'Specific files to translate (space-separated, e.g., "app common"). Leave empty for all files.'
+        required: false
+        type: string
+      languages:
+        description: 'Specific languages to translate (space-separated, e.g., "zh-Hans ja-JP"). Leave empty for all supported languages.'
+        required: false
+        type: string
+      mode:
+        description: 'Sync mode: incremental (only changes) or full (re-check all keys)'
+        required: false
+        default: 'incremental'
+        type: choice
+        options:
+          - incremental
+          - full
+
+permissions:
+  contents: write
+  pull-requests: write
+
+jobs:
+  translate:
+    if: github.repository == 'langgenius/dify'
+    runs-on: ubuntu-latest
+    timeout-minutes: 60
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v6
+        with:
+          fetch-depth: 0
+          token: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Configure Git
+        run: |
+          git config --global user.name "github-actions[bot]"
+          git config --global user.email "github-actions[bot]@users.noreply.github.com"
+
+      - name: Install pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          package_json_file: web/package.json
+          run_install: false
+
+      - name: Set up Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version: 24
+          cache: pnpm
+          cache-dependency-path: ./web/pnpm-lock.yaml
+
+      - name: Detect changed files and generate diff
+        id: detect_changes
+        run: |
+          if [ "${{ github.event_name }}" == "workflow_dispatch" ]; then
+            # Manual trigger
+            if [ -n "${{ github.event.inputs.files }}" ]; then
+              echo "CHANGED_FILES=${{ github.event.inputs.files }}" >> $GITHUB_OUTPUT
+            else
+              # Get all JSON files in en-US directory
+              files=$(ls web/i18n/en-US/*.json 2>/dev/null | xargs -n1 basename | sed 's/.json$//' | tr '\n' ' ')
+              echo "CHANGED_FILES=$files" >> $GITHUB_OUTPUT
+            fi
+            echo "TARGET_LANGS=${{ github.event.inputs.languages }}" >> $GITHUB_OUTPUT
+            echo "SYNC_MODE=${{ github.event.inputs.mode || 'incremental' }}" >> $GITHUB_OUTPUT
+
+            # For manual trigger with incremental mode, get diff from last commit
+            # For full mode, we'll do a complete check anyway
+            if [ "${{ github.event.inputs.mode }}" == "full" ]; then
+              echo "Full mode: will check all keys" > /tmp/i18n-diff.txt
+              echo "DIFF_AVAILABLE=false" >> $GITHUB_OUTPUT
+            else
+              git diff HEAD~1..HEAD -- 'web/i18n/en-US/*.json' > /tmp/i18n-diff.txt 2>/dev/null || echo "" > /tmp/i18n-diff.txt
+              if [ -s /tmp/i18n-diff.txt ]; then
+                echo "DIFF_AVAILABLE=true" >> $GITHUB_OUTPUT
+              else
+                echo "DIFF_AVAILABLE=false" >> $GITHUB_OUTPUT
+              fi
+            fi
+          elif [ "${{ github.event_name }}" == "repository_dispatch" ]; then
+            # Triggered by push via trigger-i18n-sync.yml workflow
+            # Validate required payload fields
+            if [ -z "${{ github.event.client_payload.changed_files }}" ]; then
+              echo "Error: repository_dispatch payload missing required 'changed_files' field" >&2
+              exit 1
+            fi
+            echo "CHANGED_FILES=${{ github.event.client_payload.changed_files }}" >> $GITHUB_OUTPUT
+            echo "TARGET_LANGS=" >> $GITHUB_OUTPUT
+            echo "SYNC_MODE=${{ github.event.client_payload.sync_mode || 'incremental' }}" >> $GITHUB_OUTPUT
+
+            # Decode the base64-encoded diff from the trigger workflow
+            if [ -n "${{ github.event.client_payload.diff_base64 }}" ]; then
+              if ! echo "${{ github.event.client_payload.diff_base64 }}" | base64 -d > /tmp/i18n-diff.txt 2>&1; then
+                echo "Warning: Failed to decode base64 diff payload" >&2
+                echo "" > /tmp/i18n-diff.txt
+                echo "DIFF_AVAILABLE=false" >> $GITHUB_OUTPUT
+              elif [ -s /tmp/i18n-diff.txt ]; then
+                echo "DIFF_AVAILABLE=true" >> $GITHUB_OUTPUT
+              else
+                echo "DIFF_AVAILABLE=false" >> $GITHUB_OUTPUT
+              fi
+            else
+              echo "" > /tmp/i18n-diff.txt
+              echo "DIFF_AVAILABLE=false" >> $GITHUB_OUTPUT
+            fi
+          else
+            echo "Unsupported event type: ${{ github.event_name }}"
+            exit 1
+          fi
+
+          # Truncate diff if too large (keep first 50KB)
+          if [ -f /tmp/i18n-diff.txt ]; then
+            head -c 50000 /tmp/i18n-diff.txt > /tmp/i18n-diff-truncated.txt
+            mv /tmp/i18n-diff-truncated.txt /tmp/i18n-diff.txt
+          fi
+
+          echo "Detected files: $(cat $GITHUB_OUTPUT | grep CHANGED_FILES || echo 'none')"
+
+      - name: Run Claude Code for Translation Sync
+        if: steps.detect_changes.outputs.CHANGED_FILES != ''
+        uses: anthropics/claude-code-action@v1
+        with:
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          github_token: ${{ secrets.GITHUB_TOKEN }}
+          prompt: |
+            You are a professional i18n synchronization engineer for the Dify project.
+            Your task is to keep all language translations in sync with the English source (en-US).
+
+            ## CRITICAL TOOL RESTRICTIONS
+            - Use **Read** tool to read files (NOT cat or bash)
+            - Use **Edit** tool to modify JSON files (NOT node, jq, or bash scripts)
+            - Use **Bash** ONLY for: git commands, gh commands, pnpm commands
+            - Run bash commands ONE BY ONE, never combine with && or ||
+            - NEVER use `$()` command substitution - it's not supported. Split into separate commands instead.
+
+            ## WORKING DIRECTORY & ABSOLUTE PATHS
+            Claude Code sandbox working directory may vary. Always use absolute paths:
+            - For pnpm: `pnpm --dir ${{ github.workspace }}/web <command>`
+            - For git: `git -C ${{ github.workspace }} <command>`
+            - For gh: `gh --repo ${{ github.repository }} <command>`
+            - For file paths: `${{ github.workspace }}/web/i18n/`
+
+            ## EFFICIENCY RULES
+            - **ONE Edit per language file** - batch all key additions into a single Edit
+            - Insert new keys at the beginning of JSON (after `{`), lint:fix will sort them
+            - Translate ALL keys for a language mentally first, then do ONE Edit
+
+            ## Context
+            - Changed/target files: ${{ steps.detect_changes.outputs.CHANGED_FILES }}
+            - Target languages (empty means all supported): ${{ steps.detect_changes.outputs.TARGET_LANGS }}
+            - Sync mode: ${{ steps.detect_changes.outputs.SYNC_MODE }}
+            - Translation files are located in: ${{ github.workspace }}/web/i18n/{locale}/{filename}.json
+            - Language configuration is in: ${{ github.workspace }}/web/i18n-config/languages.ts
+            - Git diff is available: ${{ steps.detect_changes.outputs.DIFF_AVAILABLE }}
+
+            ## CRITICAL DESIGN: Verify First, Then Sync
+
+            You MUST follow this three-phase approach:
+
+            ═══════════════════════════════════════════════════════════════
+            ║  PHASE 1: VERIFY - Analyze and Generate Change Report       ║
+            ═══════════════════════════════════════════════════════════════
+
+            ### Step 1.1: Analyze Git Diff (for incremental mode)
+            Use the Read tool to read `/tmp/i18n-diff.txt` to see the git diff.
+
+            Parse the diff to categorize changes:
+            - Lines with `+` (not `+++`): Added or modified values
+            - Lines with `-` (not `---`): Removed or old values
+            - Identify specific keys for each category:
+              * ADD: Keys that appear only in `+` lines (new keys)
+              * UPDATE: Keys that appear in both `-` and `+` lines (value changed)
+              * DELETE: Keys that appear only in `-` lines (removed keys)
+
+            ### Step 1.2: Read Language Configuration
+            Use the Read tool to read `${{ github.workspace }}/web/i18n-config/languages.ts`.
+            Extract all languages with `supported: true`.
+
+            ### Step 1.3: Run i18n:check for Each Language
+            ```bash
+            pnpm --dir ${{ github.workspace }}/web install --frozen-lockfile
+            ```
+            ```bash
+            pnpm --dir ${{ github.workspace }}/web run i18n:check
+            ```
+
+            This will report:
+            - Missing keys (need to ADD)
+            - Extra keys (need to DELETE)
+
+            ### Step 1.4: Generate Change Report
+
+            Create a structured report identifying:
+            ```
+            ╔══════════════════════════════════════════════════════════════╗
+            ║                    I18N SYNC CHANGE REPORT                   ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ Files to process: [list]                                     ║
+            ║ Languages to sync: [list]                                    ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ ADD (New Keys):                                              ║
+            ║   - [filename].[key]: "English value"                        ║
+            ║   ...                                                        ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ UPDATE (Modified Keys - MUST re-translate):                  ║
+            ║   - [filename].[key]: "Old value" → "New value"              ║
+            ║   ...                                                        ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ DELETE (Extra Keys):                                         ║
+            ║   - [language]/[filename].[key]                              ║
+            ║   ...                                                        ║
+            ╚══════════════════════════════════════════════════════════════╝
+            ```
+
+            **IMPORTANT**: For UPDATE detection, compare git diff to find keys where
+            the English value changed. These MUST be re-translated even if target
+            language already has a translation (it's now stale!).
+
+            ═══════════════════════════════════════════════════════════════
+            ║  PHASE 2: SYNC - Execute Changes Based on Report            ║
+            ═══════════════════════════════════════════════════════════════
+
+            ### Step 2.1: Process ADD Operations (BATCH per language file)
+
+            **CRITICAL WORKFLOW for efficiency:**
+            1. First, translate ALL new keys for ALL languages mentally
+            2. Then, for EACH language file, do ONE Edit operation:
+               - Read the file once
+               - Insert ALL new keys at the beginning (right after the opening `{`)
+               - Don't worry about alphabetical order - lint:fix will sort them later
+
+            Example Edit (adding 3 keys to zh-Hans/app.json):
+            ```
+            old_string: '{\n  "accessControl"'
+            new_string: '{\n  "newKey1": "translation1",\n  "newKey2": "translation2",\n  "newKey3": "translation3",\n  "accessControl"'
+            ```
+
+            **IMPORTANT**:
+            - ONE Edit per language file (not one Edit per key!)
+            - Always use the Edit tool. NEVER use bash scripts, node, or jq.
+
+            ### Step 2.2: Process UPDATE Operations
+
+            **IMPORTANT: Special handling for zh-Hans and ja-JP**
+            If zh-Hans or ja-JP files were ALSO modified in the same push:
+            - Run: `git -C ${{ github.workspace }} diff HEAD~1 --name-only` and check for zh-Hans or ja-JP files
+            - If found, it means someone manually translated them. Apply these rules:
+
+            1. **Missing keys**: Still ADD them (completeness required)
+            2. **Existing translations**: Compare with the NEW English value:
+               - If translation is **completely wrong** or **unrelated** → Update it
+               - If translation is **roughly correct** (captures the meaning) → Keep it, respect manual work
+               - When in doubt, **keep the manual translation**
+
+            Example:
+            - English changed: "Save" → "Save Changes"
+            - Manual translation: "保存更改" → Keep it (correct meaning)
+            - Manual translation: "删除" → Update it (completely wrong)
+
+            For other languages:
+            Use Edit tool to replace the old value with the new translation.
+            You can batch multiple updates in one Edit if they are adjacent.
+
+            ### Step 2.3: Process DELETE Operations
+            For extra keys reported by i18n:check:
+            - Run: `pnpm --dir ${{ github.workspace }}/web run i18n:check --auto-remove`
+            - Or manually remove from target language JSON files
+
+            ## Translation Guidelines
+
+            - PRESERVE all placeholders exactly as-is:
+              - `{{variable}}` - Mustache interpolation
+              - `${variable}` - Template literal
+              - `<tag>content</tag>` - HTML tags
+              - `_one`, `_other` - Pluralization suffixes (these are KEY suffixes, not values)
+            - Use appropriate language register (formal/informal) based on existing translations
+            - Match existing translation style in each language
+            - Technical terms: check existing conventions per language
+            - For CJK languages: no spaces between characters unless necessary
+            - For RTL languages (ar-TN, fa-IR): ensure proper text handling
+
+            ## Output Format Requirements
+            - Alphabetical key ordering (if original file uses it)
+            - 2-space indentation
+            - Trailing newline at end of file
+            - Valid JSON (use proper escaping for special characters)
+
+            ═══════════════════════════════════════════════════════════════
+            ║  PHASE 3: RE-VERIFY - Confirm All Issues Resolved           ║
+            ═══════════════════════════════════════════════════════════════
+
+            ### Step 3.1: Run Lint Fix (IMPORTANT!)
+            ```bash
+            pnpm --dir ${{ github.workspace }}/web lint:fix --quiet -- 'i18n/**/*.json'
+            ```
+            This ensures:
+            - JSON keys are sorted alphabetically (jsonc/sort-keys rule)
+            - Valid i18n keys (dify-i18n/valid-i18n-keys rule)
+            - No extra keys (dify-i18n/no-extra-keys rule)
+
+            ### Step 3.2: Run Final i18n Check
+            ```bash
+            pnpm --dir ${{ github.workspace }}/web run i18n:check
+            ```
+
+            ### Step 3.3: Fix Any Remaining Issues
+            If check reports issues:
+            - Go back to PHASE 2 for unresolved items
+            - Repeat until check passes
+
+            ### Step 3.4: Generate Final Summary
+            ```
+            ╔══════════════════════════════════════════════════════════════╗
+            ║                    SYNC COMPLETED SUMMARY                    ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ Language      │ Added │ Updated │ Deleted │ Status          ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ zh-Hans       │  5    │   2     │    1    │ ✓ Complete      ║
+            ║ ja-JP         │  5    │   2     │    1    │ ✓ Complete      ║
+            ║ ...           │ ...   │  ...    │   ...   │ ...             ║
+            ╠══════════════════════════════════════════════════════════════╣
+            ║ i18n:check    │ PASSED - All keys in sync                   ║
+            ╚══════════════════════════════════════════════════════════════╝
+            ```
+
+            ## Mode-Specific Behavior
+
+            **SYNC_MODE = "incremental"** (default):
+            - Focus on keys identified from git diff
+            - Also check i18n:check output for any missing/extra keys
+            - Efficient for small changes
+
+            **SYNC_MODE = "full"**:
+            - Compare ALL keys between en-US and each language
+            - Run i18n:check to identify all discrepancies
+            - Use for first-time sync or fixing historical issues
+
+            ## Important Notes
+
+            1. Always run i18n:check BEFORE and AFTER making changes
+            2. The check script is the source of truth for missing/extra keys
+            3. For UPDATE scenario: git diff is the source of truth for changed values
+            4. Create a single commit with all translation changes
+            5. If any translation fails, continue with others and report failures
+
+            ═══════════════════════════════════════════════════════════════
+            ║  PHASE 4: COMMIT AND CREATE PR                              ║
+            ═══════════════════════════════════════════════════════════════
+
+            After all translations are complete and verified:
+
+            ### Step 4.1: Check for changes
+            ```bash
+            git -C ${{ github.workspace }} status --porcelain
+            ```
+
+            If there are changes:
+
+            ### Step 4.2: Create a new branch and commit
+            Run these git commands ONE BY ONE (not combined with &&).
+            **IMPORTANT**: Do NOT use `$()` command substitution. Use two separate commands:
+
+            1. First, get the timestamp:
+            ```bash
+            date +%Y%m%d-%H%M%S
+            ```
+            (Note the output, e.g., "20260115-143052")
+
+            2. Then create branch using the timestamp value:
+            ```bash
+            git -C ${{ github.workspace }} checkout -b chore/i18n-sync-20260115-143052
+            ```
+            (Replace "20260115-143052" with the actual timestamp from step 1)
+
+            3. Stage changes:
+            ```bash
+            git -C ${{ github.workspace }} add web/i18n/
+            ```
+
+            4. Commit:
+            ```bash
+            git -C ${{ github.workspace }} commit -m "chore(i18n): sync translations with en-US - Mode: ${{ steps.detect_changes.outputs.SYNC_MODE }}"
+            ```
+
+            5. Push:
+            ```bash
+            git -C ${{ github.workspace }} push origin HEAD
+            ```
+
+            ### Step 4.3: Create Pull Request
+            ```bash
+            gh pr create --repo ${{ github.repository }} --title "chore(i18n): sync translations with en-US" --body "## Summary
+
+            This PR was automatically generated to sync i18n translation files.
+
+            ### Changes
+            - Mode: ${{ steps.detect_changes.outputs.SYNC_MODE }}
+            - Files processed: ${{ steps.detect_changes.outputs.CHANGED_FILES }}
+
+            ### Verification
+            - [x] \`i18n:check\` passed
+            - [x] \`lint:fix\` applied
+
+            🤖 Generated with Claude Code GitHub Action" --base main
+            ```
+
+          claude_args: |
+            --max-turns 150
+            --allowedTools "Read,Write,Edit,Bash(git *),Bash(git:*),Bash(gh *),Bash(gh:*),Bash(pnpm *),Bash(pnpm:*),Bash(date *),Bash(date:*),Glob,Grep"

.github/workflows/web-tests.yml

@@ -31,31 +31,15 @@ jobs:
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
-          node-version: 22
+          node-version: 24
           cache: pnpm
           cache-dependency-path: ./web/pnpm-lock.yaml
 
-      - name: Restore Jest cache
-        uses: actions/cache@v4
-        with:
-          path: web/.cache/jest
-          key: ${{ runner.os }}-jest-${{ hashFiles('web/pnpm-lock.yaml') }}
-          restore-keys: |
-            ${{ runner.os }}-jest-
-
       - name: Install dependencies
         run: pnpm install --frozen-lockfile
 
-      - name: Check i18n types synchronization
-        run: pnpm run check:i18n-types
-
       - name: Run tests
-        run: |
-          pnpm exec jest \
-            --ci \
-            --maxWorkers=100% \
-            --coverage \
-            --passWithNoTests
+        run: pnpm test:coverage
 
       - name: Coverage Summary
         if: always()
@@ -69,7 +53,7 @@ jobs:
           if [ ! -f "$COVERAGE_FILE" ] && [ ! -f "$COVERAGE_SUMMARY_FILE" ]; then
             echo "has_coverage=false" >> "$GITHUB_OUTPUT"
             echo "### 🚨 Test Coverage Report :test_tube:" >> "$GITHUB_STEP_SUMMARY"
-            echo "Coverage data not found. Ensure Jest runs with coverage enabled." >> "$GITHUB_STEP_SUMMARY"
+            echo "Coverage data not found. Ensure Vitest runs with coverage enabled." >> "$GITHUB_STEP_SUMMARY"
             exit 0
           fi
 
@@ -365,7 +349,7 @@ jobs:
               .join(' | ')} |`;
 
             console.log('');
-            console.log('<details><summary>Jest coverage table</summary>');
+            console.log('<details><summary>Vitest coverage table</summary>');
             console.log('');
             console.log(headerRow);
             console.log(dividerRow);

.gitignore

@@ -139,7 +139,6 @@ pyrightconfig.json
 .idea/'
 
 .DS_Store
-web/.vscode/settings.json
 
 # Intellij IDEA Files
 .idea/*
@@ -196,6 +195,7 @@ docker/nginx/ssl/*
 !docker/nginx/ssl/.gitkeep
 docker/middleware.env
 docker/docker-compose.override.yaml
+docker/env-backup/*
 
 sdks/python-client/build
 sdks/python-client/dist
@@ -205,7 +205,6 @@ sdks/python-client/dify_client.egg-info
 !.vscode/launch.json.template
 !.vscode/README.md
 api/.vscode
-web/.vscode
 # vscode Code History Extension
 .history
 
@@ -220,15 +219,6 @@ plugins.jsonl
 # mise
 mise.toml
 
-# Next.js build output
-.next/
-
-# PWA generated files
-web/public/sw.js
-web/public/sw.js.map
-web/public/workbox-*.js
-web/public/workbox-*.js.map
-web/public/fallback-*.js
 
 # AI Assistant
 .roo/

api/.env.example

@@ -116,6 +116,7 @@ ALIYUN_OSS_AUTH_VERSION=v1
 ALIYUN_OSS_REGION=your-region
 # Don't start with '/'. OSS doesn't support leading slash in object names.
 ALIYUN_OSS_PATH=your-path
+ALIYUN_CLOUDBOX_ID=your-cloudbox-id
 
 # Google Storage configuration
 GOOGLE_STORAGE_BUCKET_NAME=your-bucket-name
@@ -133,6 +134,7 @@ HUAWEI_OBS_BUCKET_NAME=your-bucket-name
 HUAWEI_OBS_SECRET_KEY=your-secret-key
 HUAWEI_OBS_ACCESS_KEY=your-access-key
 HUAWEI_OBS_SERVER=your-server-url
+HUAWEI_OBS_PATH_STYLE=false
 
 # Baidu OBS Storage Configuration
 BAIDU_OBS_BUCKET_NAME=your-bucket-name
@@ -690,7 +692,6 @@ ANNOTATION_IMPORT_RATE_LIMIT_PER_MINUTE=5
 ANNOTATION_IMPORT_RATE_LIMIT_PER_HOUR=20
 # Maximum number of concurrent annotation import tasks per tenant
 ANNOTATION_IMPORT_MAX_CONCURRENT=5
-
 # Sandbox expired records clean configuration
 SANDBOX_EXPIRED_RECORDS_CLEAN_GRACEFUL_PERIOD=21
 SANDBOX_EXPIRED_RECORDS_CLEAN_BATCH_SIZE=1000

api/Dockerfile

@@ -50,16 +50,33 @@ WORKDIR /app/api
 
 # Create non-root user
 ARG dify_uid=1001
+ARG NODE_MAJOR=22
+ARG NODE_PACKAGE_VERSION=22.21.0-1nodesource1
+ARG NODESOURCE_KEY_FPR=6F71F525282841EEDAF851B42F59B5F99B1BE0B4
 RUN groupadd -r -g ${dify_uid} dify && \
     useradd -r -u ${dify_uid} -g ${dify_uid} -s /bin/bash dify && \
     chown -R dify:dify /app
 
 RUN \
     apt-get update \
+    && apt-get install -y --no-install-recommends \
+        ca-certificates \
+        curl \
+        gnupg \
+    && mkdir -p /etc/apt/keyrings \
+    && curl -fsSL https://deb.nodesource.com/gpgkey/nodesource-repo.gpg.key -o /tmp/nodesource.gpg \
+    && gpg --show-keys --with-colons /tmp/nodesource.gpg \
+        | awk -F: '/^fpr:/ {print $10}' \
+        | grep -Fx "${NODESOURCE_KEY_FPR}" \
+    && gpg --dearmor -o /etc/apt/keyrings/nodesource.gpg /tmp/nodesource.gpg \
+    && rm -f /tmp/nodesource.gpg \
+    && echo "deb [signed-by=/etc/apt/keyrings/nodesource.gpg] https://deb.nodesource.com/node_${NODE_MAJOR}.x nodistro main" \
+        > /etc/apt/sources.list.d/nodesource.list \
+    && apt-get update \
     # Install dependencies
     && apt-get install -y --no-install-recommends \
         # basic environment
-        curl nodejs \
+        nodejs=${NODE_PACKAGE_VERSION} \
         # for gmpy2 \
         libgmp-dev libmpfr-dev libmpc-dev \
         # For Security

api/configs/middleware/storage/aliyun_oss_storage_config.py

@@ -41,3 +41,8 @@ class AliyunOSSStorageConfig(BaseSettings):
         description="Base path within the bucket to store objects (e.g., 'my-app-data/')",
         default=None,
     )
+
+    ALIYUN_CLOUDBOX_ID: str | None = Field(
+        description="Cloudbox id for aliyun cloudbox service",
+        default=None,
+    )

api/configs/middleware/storage/huawei_obs_storage_config.py

@@ -26,3 +26,8 @@ class HuaweiCloudOBSStorageConfig(BaseSettings):
         description="Endpoint URL for Huawei Cloud OBS (e.g., 'https://obs.cn-north-4.myhuaweicloud.com')",
         default=None,
     )
+
+    HUAWEI_OBS_PATH_STYLE: bool = Field(
+        description="Flag to indicate whether to use path-style URLs for OBS requests",
+        default=False,
+    )

api/controllers/common/file_response.py

@@ -0,0 +1,57 @@
+import os
+from email.message import Message
+from urllib.parse import quote
+
+from flask import Response
+
+HTML_MIME_TYPES = frozenset({"text/html", "application/xhtml+xml"})
+HTML_EXTENSIONS = frozenset({"html", "htm"})
+
+
+def _normalize_mime_type(mime_type: str | None) -> str:
+    if not mime_type:
+        return ""
+    message = Message()
+    message["Content-Type"] = mime_type
+    return message.get_content_type().strip().lower()
+
+
+def _is_html_extension(extension: str | None) -> bool:
+    if not extension:
+        return False
+    return extension.lstrip(".").lower() in HTML_EXTENSIONS
+
+
+def is_html_content(mime_type: str | None, filename: str | None, extension: str | None = None) -> bool:
+    normalized_mime_type = _normalize_mime_type(mime_type)
+    if normalized_mime_type in HTML_MIME_TYPES:
+        return True
+
+    if _is_html_extension(extension):
+        return True
+
+    if filename:
+        return _is_html_extension(os.path.splitext(filename)[1])
+
+    return False
+
+
+def enforce_download_for_html(
+    response: Response,
+    *,
+    mime_type: str | None,
+    filename: str | None,
+    extension: str | None = None,
+) -> bool:
+    if not is_html_content(mime_type, filename, extension):
+        return False
+
+    if filename:
+        encoded_filename = quote(filename)
+        response.headers["Content-Disposition"] = f"attachment; filename*=UTF-8''{encoded_filename}"
+    else:
+        response.headers["Content-Disposition"] = "attachment"
+
+    response.headers["Content-Type"] = "application/octet-stream"
+    response.headers["X-Content-Type-Options"] = "nosniff"
+    return True

api/controllers/console/auth/activate.py

@@ -7,9 +7,9 @@ from controllers.console import console_ns
 from controllers.console.error import AlreadyActivateError
 from extensions.ext_database import db
 from libs.datetime_utils import naive_utc_now
-from libs.helper import EmailStr, extract_remote_ip, timezone
+from libs.helper import EmailStr, timezone
 from models import AccountStatus
-from services.account_service import AccountService, RegisterService
+from services.account_service import RegisterService
 
 DEFAULT_REF_TEMPLATE_SWAGGER_2_0 = "#/definitions/{model}"
 
@@ -93,7 +93,6 @@ class ActivateApi(Resource):
             "ActivationResponse",
             {
                 "result": fields.String(description="Operation result"),
-                "data": fields.Raw(description="Login token data"),
             },
         ),
     )
@@ -117,6 +116,4 @@ class ActivateApi(Resource):
         account.initialized_at = naive_utc_now()
         db.session.commit()
 
-        token_pair = AccountService.login(account, ip_address=extract_remote_ip(request))
-
-        return {"result": "success", "data": token_pair.model_dump()}
+        return {"result": "success"}

api/controllers/console/datasets/datasets_document.py

@@ -572,7 +572,7 @@ class DocumentBatchIndexingEstimateApi(DocumentResource):
                     datasource_type=DatasourceType.NOTION,
                     notion_info=NotionInfo.model_validate(
                         {
-                            "credential_id": data_source_info["credential_id"],
+                            "credential_id": data_source_info.get("credential_id"),
                             "notion_workspace_id": data_source_info["notion_workspace_id"],
                             "notion_obj_id": data_source_info["notion_page_id"],
                             "notion_page_type": data_source_info["type"],

api/controllers/console/explore/conversation.py

@@ -1,5 +1,4 @@
 from typing import Any
-from uuid import UUID
 
 from flask import request
 from flask_restx import marshal_with
@@ -13,6 +12,7 @@ from controllers.console.explore.wraps import InstalledAppResource
 from core.app.entities.app_invoke_entities import InvokeFrom
 from extensions.ext_database import db
 from fields.conversation_fields import conversation_infinite_scroll_pagination_fields, simple_conversation_fields
+from libs.helper import UUIDStrOrEmpty
 from libs.login import current_user
 from models import Account
 from models.model import AppMode
@@ -24,7 +24,7 @@ from .. import console_ns
 
 
 class ConversationListQuery(BaseModel):
-    last_id: UUID | None = None
+    last_id: UUIDStrOrEmpty | None = None
     limit: int = Field(default=20, ge=1, le=100)
     pinned: bool | None = None
 

api/controllers/console/explore/installed_app.py

@@ -2,7 +2,8 @@ import logging
 from typing import Any
 
 from flask import request
-from flask_restx import Resource, inputs, marshal_with, reqparse
+from flask_restx import Resource, marshal_with
+from pydantic import BaseModel
 from sqlalchemy import and_, select
 from werkzeug.exceptions import BadRequest, Forbidden, NotFound
 
@@ -18,6 +19,15 @@ from services.account_service import TenantService
 from services.enterprise.enterprise_service import EnterpriseService
 from services.feature_service import FeatureService
 
+
+class InstalledAppCreatePayload(BaseModel):
+    app_id: str
+
+
+class InstalledAppUpdatePayload(BaseModel):
+    is_pinned: bool | None = None
+
+
 logger = logging.getLogger(__name__)
 
 
@@ -105,26 +115,25 @@ class InstalledAppsListApi(Resource):
     @account_initialization_required
     @cloud_edition_billing_resource_check("apps")
     def post(self):
-        parser = reqparse.RequestParser().add_argument("app_id", type=str, required=True, help="Invalid app_id")
-        args = parser.parse_args()
+        payload = InstalledAppCreatePayload.model_validate(console_ns.payload or {})
 
-        recommended_app = db.session.query(RecommendedApp).where(RecommendedApp.app_id == args["app_id"]).first()
+        recommended_app = db.session.query(RecommendedApp).where(RecommendedApp.app_id == payload.app_id).first()
         if recommended_app is None:
-            raise NotFound("App not found")
+            raise NotFound("Recommended app not found")
 
         _, current_tenant_id = current_account_with_tenant()
 
-        app = db.session.query(App).where(App.id == args["app_id"]).first()
+        app = db.session.query(App).where(App.id == payload.app_id).first()
 
         if app is None:
-            raise NotFound("App not found")
+            raise NotFound("App entity not found")
 
         if not app.is_public:
             raise Forbidden("You can't install a non-public app")
 
         installed_app = (
             db.session.query(InstalledApp)
-            .where(and_(InstalledApp.app_id == args["app_id"], InstalledApp.tenant_id == current_tenant_id))
+            .where(and_(InstalledApp.app_id == payload.app_id, InstalledApp.tenant_id == current_tenant_id))
             .first()
         )
 
@@ -133,7 +142,7 @@ class InstalledAppsListApi(Resource):
             recommended_app.install_count += 1
 
             new_installed_app = InstalledApp(
-                app_id=args["app_id"],
+                app_id=payload.app_id,
                 tenant_id=current_tenant_id,
                 app_owner_tenant_id=app.tenant_id,
                 is_pinned=False,
@@ -163,12 +172,11 @@ class InstalledAppApi(InstalledAppResource):
         return {"result": "success", "message": "App uninstalled successfully"}, 204
 
     def patch(self, installed_app):
-        parser = reqparse.RequestParser().add_argument("is_pinned", type=inputs.boolean)
-        args = parser.parse_args()
+        payload = InstalledAppUpdatePayload.model_validate(console_ns.payload or {})
 
         commit_args = False
-        if "is_pinned" in args:
-            installed_app.is_pinned = args["is_pinned"]
+        if payload.is_pinned is not None:
+            installed_app.is_pinned = payload.is_pinned
             commit_args = True
 
         if commit_args:

api/controllers/console/extension.py

@@ -1,14 +1,32 @@
-from flask_restx import Resource, fields, marshal_with, reqparse
+from flask import request
+from flask_restx import Resource, fields, marshal_with
+from pydantic import BaseModel, Field
 
 from constants import HIDDEN_VALUE
-from controllers.console import console_ns
-from controllers.console.wraps import account_initialization_required, setup_required
 from fields.api_based_extension_fields import api_based_extension_fields
 from libs.login import current_account_with_tenant, login_required
 from models.api_based_extension import APIBasedExtension
 from services.api_based_extension_service import APIBasedExtensionService
 from services.code_based_extension_service import CodeBasedExtensionService
 
+from ..common.schema import register_schema_models
+from . import console_ns
+from .wraps import account_initialization_required, setup_required
+
+
+class CodeBasedExtensionQuery(BaseModel):
+    module: str
+
+
+class APIBasedExtensionPayload(BaseModel):
+    name: str = Field(description="Extension name")
+    api_endpoint: str = Field(description="API endpoint URL")
+    api_key: str = Field(description="API key for authentication")
+
+
+register_schema_models(console_ns, APIBasedExtensionPayload)
+
+
 api_based_extension_model = console_ns.model("ApiBasedExtensionModel", api_based_extension_fields)
 
 api_based_extension_list_model = fields.List(fields.Nested(api_based_extension_model))
@@ -18,11 +36,7 @@ api_based_extension_list_model = fields.List(fields.Nested(api_based_extension_m
 class CodeBasedExtensionAPI(Resource):
     @console_ns.doc("get_code_based_extension")
     @console_ns.doc(description="Get code-based extension data by module name")
-    @console_ns.expect(
-        console_ns.parser().add_argument(
-            "module", type=str, required=True, location="args", help="Extension module name"
-        )
-    )
+    @console_ns.doc(params={"module": "Extension module name"})
     @console_ns.response(
         200,
         "Success",
@@ -35,10 +49,9 @@ class CodeBasedExtensionAPI(Resource):
     @login_required
     @account_initialization_required
     def get(self):
-        parser = reqparse.RequestParser().add_argument("module", type=str, required=True, location="args")
-        args = parser.parse_args()
+        query = CodeBasedExtensionQuery.model_validate(request.args.to_dict(flat=True))  # type: ignore
 
-        return {"module": args["module"], "data": CodeBasedExtensionService.get_code_based_extension(args["module"])}
+        return {"module": query.module, "data": CodeBasedExtensionService.get_code_based_extension(query.module)}
 
 
 @console_ns.route("/api-based-extension")
@@ -56,30 +69,21 @@ class APIBasedExtensionAPI(Resource):
 
     @console_ns.doc("create_api_based_extension")
     @console_ns.doc(description="Create a new API-based extension")
-    @console_ns.expect(
-        console_ns.model(
-            "CreateAPIBasedExtensionRequest",
-            {
-                "name": fields.String(required=True, description="Extension name"),
-                "api_endpoint": fields.String(required=True, description="API endpoint URL"),
-                "api_key": fields.String(required=True, description="API key for authentication"),
-            },
-        )
-    )
+    @console_ns.expect(console_ns.models[APIBasedExtensionPayload.__name__])
     @console_ns.response(201, "Extension created successfully", api_based_extension_model)
     @setup_required
     @login_required
     @account_initialization_required
     @marshal_with(api_based_extension_model)
     def post(self):
-        args = console_ns.payload
+        payload = APIBasedExtensionPayload.model_validate(console_ns.payload or {})
         _, current_tenant_id = current_account_with_tenant()
 
         extension_data = APIBasedExtension(
             tenant_id=current_tenant_id,
-            name=args["name"],
-            api_endpoint=args["api_endpoint"],
-            api_key=args["api_key"],
+            name=payload.name,
+            api_endpoint=payload.api_endpoint,
+            api_key=payload.api_key,
         )
 
         return APIBasedExtensionService.save(extension_data)
@@ -104,16 +108,7 @@ class APIBasedExtensionDetailAPI(Resource):
     @console_ns.doc("update_api_based_extension")
     @console_ns.doc(description="Update API-based extension")
     @console_ns.doc(params={"id": "Extension ID"})
-    @console_ns.expect(
-        console_ns.model(
-            "UpdateAPIBasedExtensionRequest",
-            {
-                "name": fields.String(required=True, description="Extension name"),
-                "api_endpoint": fields.String(required=True, description="API endpoint URL"),
-                "api_key": fields.String(required=True, description="API key for authentication"),
-            },
-        )
-    )
+    @console_ns.expect(console_ns.models[APIBasedExtensionPayload.__name__])
     @console_ns.response(200, "Extension updated successfully", api_based_extension_model)
     @setup_required
     @login_required
@@ -125,13 +120,13 @@ class APIBasedExtensionDetailAPI(Resource):
 
         extension_data_from_db = APIBasedExtensionService.get_with_tenant_id(current_tenant_id, api_based_extension_id)
 
-        args = console_ns.payload
+        payload = APIBasedExtensionPayload.model_validate(console_ns.payload or {})
 
-        extension_data_from_db.name = args["name"]
-        extension_data_from_db.api_endpoint = args["api_endpoint"]
+        extension_data_from_db.name = payload.name
+        extension_data_from_db.api_endpoint = payload.api_endpoint
 
-        if args["api_key"] != HIDDEN_VALUE:
-            extension_data_from_db.api_key = args["api_key"]
+        if payload.api_key != HIDDEN_VALUE:
+            extension_data_from_db.api_key = payload.api_key
 
         return APIBasedExtensionService.save(extension_data_from_db)
 

api/controllers/console/workspace/plugin.py

@@ -1,5 +1,6 @@
 import io
-from typing import Literal
+from collections.abc import Mapping
+from typing import Any, Literal
 
 from flask import request, send_file
 from flask_restx import Resource
@@ -141,6 +142,15 @@ class ParserDynamicOptions(BaseModel):
     provider_type: Literal["tool", "trigger"]
 
 
+class ParserDynamicOptionsWithCredentials(BaseModel):
+    plugin_id: str
+    provider: str
+    action: str
+    parameter: str
+    credential_id: str
+    credentials: Mapping[str, Any]
+
+
 class PluginPermissionSettingsPayload(BaseModel):
     install_permission: TenantPluginPermission.InstallPermission = TenantPluginPermission.InstallPermission.EVERYONE
     debug_permission: TenantPluginPermission.DebugPermission = TenantPluginPermission.DebugPermission.EVERYONE
@@ -183,6 +193,7 @@ reg(ParserGithubUpgrade)
 reg(ParserUninstall)
 reg(ParserPermissionChange)
 reg(ParserDynamicOptions)
+reg(ParserDynamicOptionsWithCredentials)
 reg(ParserPreferencesChange)
 reg(ParserExcludePlugin)
 reg(ParserReadme)
@@ -657,6 +668,37 @@ class PluginFetchDynamicSelectOptionsApi(Resource):
         return jsonable_encoder({"options": options})
 
 
+@console_ns.route("/workspaces/current/plugin/parameters/dynamic-options-with-credentials")
+class PluginFetchDynamicSelectOptionsWithCredentialsApi(Resource):
+    @console_ns.expect(console_ns.models[ParserDynamicOptionsWithCredentials.__name__])
+    @setup_required
+    @login_required
+    @is_admin_or_owner_required
+    @account_initialization_required
+    def post(self):
+        """Fetch dynamic options using credentials directly (for edit mode)."""
+        current_user, tenant_id = current_account_with_tenant()
+        user_id = current_user.id
+
+        args = ParserDynamicOptionsWithCredentials.model_validate(console_ns.payload)
+
+        try:
+            options = PluginParameterService.get_dynamic_select_options_with_credentials(
+                tenant_id=tenant_id,
+                user_id=user_id,
+                plugin_id=args.plugin_id,
+                provider=args.provider,
+                action=args.action,
+                parameter=args.parameter,
+                credential_id=args.credential_id,
+                credentials=args.credentials,
+            )
+        except PluginDaemonClientSideError as e:
+            raise ValueError(e)
+
+        return jsonable_encoder({"options": options})
+
+
 @console_ns.route("/workspaces/current/plugin/preferences/change")
 class PluginChangePreferencesApi(Resource):
     @console_ns.expect(console_ns.models[ParserPreferencesChange.__name__])

api/controllers/console/workspace/tool_providers.py

@@ -18,6 +18,7 @@ from controllers.console.wraps import (
     setup_required,
 )
 from core.entities.mcp_provider import MCPAuthentication, MCPConfiguration
+from core.helper.tool_provider_cache import ToolProviderListCache
 from core.mcp.auth.auth_flow import auth, handle_callback
 from core.mcp.error import MCPAuthError, MCPError, MCPRefreshTokenError
 from core.mcp.mcp_client import MCPClient
@@ -944,7 +945,7 @@ class ToolProviderMCPApi(Resource):
         configuration = MCPConfiguration.model_validate(args["configuration"])
         authentication = MCPAuthentication.model_validate(args["authentication"]) if args["authentication"] else None
 
-        # Create provider
+        # Create provider in transaction
         with Session(db.engine) as session, session.begin():
             service = MCPToolManageService(session=session)
             result = service.create_provider(
@@ -960,7 +961,11 @@ class ToolProviderMCPApi(Resource):
                 configuration=configuration,
                 authentication=authentication,
             )
-            return jsonable_encoder(result)
+
+        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
+        ToolProviderListCache.invalidate_cache(tenant_id)
+
+        return jsonable_encoder(result)
 
     @console_ns.expect(parser_mcp_put)
     @setup_required
@@ -972,17 +977,23 @@ class ToolProviderMCPApi(Resource):
         authentication = MCPAuthentication.model_validate(args["authentication"]) if args["authentication"] else None
         _, current_tenant_id = current_account_with_tenant()
 
-        # Step 1: Validate server URL change if needed (includes URL format validation and network operation)
-        validation_result = None
+        # Step 1: Get provider data for URL validation (short-lived session, no network I/O)
+        validation_data = None
         with Session(db.engine) as session:
             service = MCPToolManageService(session=session)
-            validation_result = service.validate_server_url_change(
-                tenant_id=current_tenant_id, provider_id=args["provider_id"], new_server_url=args["server_url"]
+            validation_data = service.get_provider_for_url_validation(
+                tenant_id=current_tenant_id, provider_id=args["provider_id"]
             )
 
-            # No need to check for errors here, exceptions will be raised directly
+        # Step 2: Perform URL validation with network I/O OUTSIDE of any database session
+        # This prevents holding database locks during potentially slow network operations
+        validation_result = MCPToolManageService.validate_server_url_standalone(
+            tenant_id=current_tenant_id,
+            new_server_url=args["server_url"],
+            validation_data=validation_data,
+        )
 
-        # Step 2: Perform database update in a transaction
+        # Step 3: Perform database update in a transaction
         with Session(db.engine) as session, session.begin():
             service = MCPToolManageService(session=session)
             service.update_provider(
@@ -999,7 +1010,11 @@ class ToolProviderMCPApi(Resource):
                 authentication=authentication,
                 validation_result=validation_result,
             )
-            return {"result": "success"}
+
+        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
+        ToolProviderListCache.invalidate_cache(current_tenant_id)
+
+        return {"result": "success"}
 
     @console_ns.expect(parser_mcp_delete)
     @setup_required
@@ -1012,7 +1027,11 @@ class ToolProviderMCPApi(Resource):
         with Session(db.engine) as session, session.begin():
             service = MCPToolManageService(session=session)
             service.delete_provider(tenant_id=current_tenant_id, provider_id=args["provider_id"])
-            return {"result": "success"}
+
+        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
+        ToolProviderListCache.invalidate_cache(current_tenant_id)
+
+        return {"result": "success"}
 
 
 parser_auth = (
@@ -1062,6 +1081,8 @@ class ToolMCPAuthApi(Resource):
                         credentials=provider_entity.credentials,
                         authed=True,
                     )
+                # Invalidate cache after updating credentials
+                ToolProviderListCache.invalidate_cache(tenant_id)
                 return {"result": "success"}
         except MCPAuthError as e:
             try:
@@ -1075,16 +1096,22 @@ class ToolMCPAuthApi(Resource):
                 with Session(db.engine) as session, session.begin():
                     service = MCPToolManageService(session=session)
                     response = service.execute_auth_actions(auth_result)
+                    # Invalidate cache after auth actions may have updated provider state
+                    ToolProviderListCache.invalidate_cache(tenant_id)
                     return response
             except MCPRefreshTokenError as e:
                 with Session(db.engine) as session, session.begin():
                     service = MCPToolManageService(session=session)
                     service.clear_provider_credentials(provider_id=provider_id, tenant_id=tenant_id)
+                # Invalidate cache after clearing credentials
+                ToolProviderListCache.invalidate_cache(tenant_id)
                 raise ValueError(f"Failed to refresh token, please try to authorize again: {e}") from e
         except (MCPError, ValueError) as e:
             with Session(db.engine) as session, session.begin():
                 service = MCPToolManageService(session=session)
                 service.clear_provider_credentials(provider_id=provider_id, tenant_id=tenant_id)
+            # Invalidate cache after clearing credentials
+            ToolProviderListCache.invalidate_cache(tenant_id)
             raise ValueError(f"Failed to connect to MCP server: {e}") from e
 
 

api/controllers/console/workspace/trigger_providers.py

@@ -1,11 +1,15 @@
 import logging
+from collections.abc import Mapping
+from typing import Any
 
 from flask import make_response, redirect, request
 from flask_restx import Resource, reqparse
+from pydantic import BaseModel, Field
 from sqlalchemy.orm import Session
 from werkzeug.exceptions import BadRequest, Forbidden
 
 from configs import dify_config
+from constants import HIDDEN_VALUE, UNKNOWN_VALUE
 from controllers.web.error import NotFoundError
 from core.model_runtime.utils.encoders import jsonable_encoder
 from core.plugin.entities.plugin_daemon import CredentialType
@@ -32,6 +36,32 @@ from ..wraps import (
 logger = logging.getLogger(__name__)
 
 
+class TriggerSubscriptionUpdateRequest(BaseModel):
+    """Request payload for updating a trigger subscription"""
+
+    name: str | None = Field(default=None, description="The name for the subscription")
+    credentials: Mapping[str, Any] | None = Field(default=None, description="The credentials for the subscription")
+    parameters: Mapping[str, Any] | None = Field(default=None, description="The parameters for the subscription")
+    properties: Mapping[str, Any] | None = Field(default=None, description="The properties for the subscription")
+
+
+class TriggerSubscriptionVerifyRequest(BaseModel):
+    """Request payload for verifying subscription credentials."""
+
+    credentials: Mapping[str, Any] = Field(description="The credentials to verify")
+
+
+console_ns.schema_model(
+    TriggerSubscriptionUpdateRequest.__name__,
+    TriggerSubscriptionUpdateRequest.model_json_schema(ref_template="#/definitions/{model}"),
+)
+
+console_ns.schema_model(
+    TriggerSubscriptionVerifyRequest.__name__,
+    TriggerSubscriptionVerifyRequest.model_json_schema(ref_template="#/definitions/{model}"),
+)
+
+
 @console_ns.route("/workspaces/current/trigger-provider/<path:provider>/icon")
 class TriggerProviderIconApi(Resource):
     @setup_required
@@ -155,16 +185,16 @@ parser_api = (
 
 
 @console_ns.route(
-    "/workspaces/current/trigger-provider/<path:provider>/subscriptions/builder/verify/<path:subscription_builder_id>",
+    "/workspaces/current/trigger-provider/<path:provider>/subscriptions/builder/verify-and-update/<path:subscription_builder_id>",
 )
-class TriggerSubscriptionBuilderVerifyApi(Resource):
+class TriggerSubscriptionBuilderVerifyAndUpdateApi(Resource):
     @console_ns.expect(parser_api)
     @setup_required
     @login_required
     @edit_permission_required
     @account_initialization_required
     def post(self, provider, subscription_builder_id):
-        """Verify a subscription instance for a trigger provider"""
+        """Verify and update a subscription instance for a trigger provider"""
         user = current_user
         assert user.current_tenant_id is not None
 
@@ -289,6 +319,83 @@ class TriggerSubscriptionBuilderBuildApi(Resource):
             raise ValueError(str(e)) from e
 
 
+@console_ns.route(
+    "/workspaces/current/trigger-provider/<path:subscription_id>/subscriptions/update",
+)
+class TriggerSubscriptionUpdateApi(Resource):
+    @console_ns.expect(console_ns.models[TriggerSubscriptionUpdateRequest.__name__])
+    @setup_required
+    @login_required
+    @edit_permission_required
+    @account_initialization_required
+    def post(self, subscription_id: str):
+        """Update a subscription instance"""
+        user = current_user
+        assert user.current_tenant_id is not None
+
+        args = TriggerSubscriptionUpdateRequest.model_validate(console_ns.payload)
+
+        subscription = TriggerProviderService.get_subscription_by_id(
+            tenant_id=user.current_tenant_id,
+            subscription_id=subscription_id,
+        )
+        if not subscription:
+            raise NotFoundError(f"Subscription {subscription_id} not found")
+
+        provider_id = TriggerProviderID(subscription.provider_id)
+
+        try:
+            # rename only
+            if (
+                args.name is not None
+                and args.credentials is None
+                and args.parameters is None
+                and args.properties is None
+            ):
+                TriggerProviderService.update_trigger_subscription(
+                    tenant_id=user.current_tenant_id,
+                    subscription_id=subscription_id,
+                    name=args.name,
+                )
+                return 200
+
+            # rebuild for create automatically by the provider
+            match subscription.credential_type:
+                case CredentialType.UNAUTHORIZED:
+                    TriggerProviderService.update_trigger_subscription(
+                        tenant_id=user.current_tenant_id,
+                        subscription_id=subscription_id,
+                        name=args.name,
+                        properties=args.properties,
+                    )
+                    return 200
+                case CredentialType.API_KEY | CredentialType.OAUTH2:
+                    if args.credentials:
+                        new_credentials: dict[str, Any] = {
+                            key: value if value != HIDDEN_VALUE else subscription.credentials.get(key, UNKNOWN_VALUE)
+                            for key, value in args.credentials.items()
+                        }
+                    else:
+                        new_credentials = subscription.credentials
+
+                    TriggerProviderService.rebuild_trigger_subscription(
+                        tenant_id=user.current_tenant_id,
+                        name=args.name,
+                        provider_id=provider_id,
+                        subscription_id=subscription_id,
+                        credentials=new_credentials,
+                        parameters=args.parameters or subscription.parameters,
+                    )
+                    return 200
+                case _:
+                    raise BadRequest("Invalid credential type")
+        except ValueError as e:
+            raise BadRequest(str(e))
+        except Exception as e:
+            logger.exception("Error updating subscription", exc_info=e)
+            raise
+
+
 @console_ns.route(
     "/workspaces/current/trigger-provider/<path:subscription_id>/subscriptions/delete",
 )
@@ -576,3 +683,38 @@ class TriggerOAuthClientManageApi(Resource):
         except Exception as e:
             logger.exception("Error removing OAuth client", exc_info=e)
             raise
+
+
+@console_ns.route(
+    "/workspaces/current/trigger-provider/<path:provider>/subscriptions/verify/<path:subscription_id>",
+)
+class TriggerSubscriptionVerifyApi(Resource):
+    @console_ns.expect(console_ns.models[TriggerSubscriptionVerifyRequest.__name__])
+    @setup_required
+    @login_required
+    @edit_permission_required
+    @account_initialization_required
+    def post(self, provider, subscription_id):
+        """Verify credentials for an existing subscription (edit mode only)"""
+        user = current_user
+        assert user.current_tenant_id is not None
+
+        verify_request: TriggerSubscriptionVerifyRequest = TriggerSubscriptionVerifyRequest.model_validate(
+            console_ns.payload
+        )
+
+        try:
+            result = TriggerProviderService.verify_subscription_credentials(
+                tenant_id=user.current_tenant_id,
+                user_id=user.id,
+                provider_id=TriggerProviderID(provider),
+                subscription_id=subscription_id,
+                credentials=verify_request.credentials,
+            )
+            return result
+        except ValueError as e:
+            logger.warning("Credential verification failed", exc_info=e)
+            raise BadRequest(str(e)) from e
+        except Exception as e:
+            logger.exception("Error verifying subscription credentials", exc_info=e)
+            raise BadRequest(str(e)) from e

api/controllers/files/image_preview.py

@@ -7,6 +7,7 @@ from werkzeug.exceptions import NotFound
 
 import services
 from controllers.common.errors import UnsupportedFileTypeError
+from controllers.common.file_response import enforce_download_for_html
 from controllers.files import files_ns
 from extensions.ext_database import db
 from services.account_service import TenantService
@@ -138,6 +139,13 @@ class FilePreviewApi(Resource):
             response.headers["Content-Disposition"] = f"attachment; filename*=UTF-8''{encoded_filename}"
             response.headers["Content-Type"] = "application/octet-stream"
 
+        enforce_download_for_html(
+            response,
+            mime_type=upload_file.mime_type,
+            filename=upload_file.name,
+            extension=upload_file.extension,
+        )
+
         return response
 
 

api/controllers/files/tool_files.py

@@ -6,6 +6,7 @@ from pydantic import BaseModel, Field
 from werkzeug.exceptions import Forbidden, NotFound
 
 from controllers.common.errors import UnsupportedFileTypeError
+from controllers.common.file_response import enforce_download_for_html
 from controllers.files import files_ns
 from core.tools.signature import verify_tool_file_signature
 from core.tools.tool_file_manager import ToolFileManager
@@ -78,4 +79,11 @@ class ToolFileApi(Resource):
             encoded_filename = quote(tool_file.name)
             response.headers["Content-Disposition"] = f"attachment; filename*=UTF-8''{encoded_filename}"
 
+        enforce_download_for_html(
+            response,
+            mime_type=tool_file.mimetype,
+            filename=tool_file.name,
+            extension=extension,
+        )
+
         return response

api/controllers/service_api/app/conversation.py

@@ -4,7 +4,7 @@ from uuid import UUID
 from flask import request
 from flask_restx import Resource
 from flask_restx._http import HTTPStatus
-from pydantic import BaseModel, Field, model_validator
+from pydantic import BaseModel, Field, field_validator, model_validator
 from sqlalchemy.orm import Session
 from werkzeug.exceptions import BadRequest, NotFound
 
@@ -51,6 +51,32 @@ class ConversationRenamePayload(BaseModel):
 class ConversationVariablesQuery(BaseModel):
     last_id: UUID | None = Field(default=None, description="Last variable ID for pagination")
     limit: int = Field(default=20, ge=1, le=100, description="Number of variables to return")
+    variable_name: str | None = Field(
+        default=None, description="Filter variables by name", min_length=1, max_length=255
+    )
+
+    @field_validator("variable_name", mode="before")
+    @classmethod
+    def validate_variable_name(cls, v: str | None) -> str | None:
+        """
+        Validate variable_name to prevent injection attacks.
+        """
+        if v is None:
+            return v
+
+        # Only allow safe characters: alphanumeric, underscore, hyphen, period
+        if not v.replace("-", "").replace("_", "").replace(".", "").isalnum():
+            raise ValueError(
+                "Variable name can only contain letters, numbers, hyphens (-), underscores (_), and periods (.)"
+            )
+
+        # Prevent SQL injection patterns
+        dangerous_patterns = ["'", '"', ";", "--", "/*", "*/", "xp_", "sp_"]
+        for pattern in dangerous_patterns:
+            if pattern in v.lower():
+                raise ValueError(f"Variable name contains invalid characters: {pattern}")
+
+        return v
 
 
 class ConversationVariableUpdatePayload(BaseModel):
@@ -199,7 +225,7 @@ class ConversationVariablesApi(Resource):
 
         try:
             return ConversationService.get_conversational_variable(
-                app_model, conversation_id, end_user, query_args.limit, last_id
+                app_model, conversation_id, end_user, query_args.limit, last_id, query_args.variable_name
             )
         except services.errors.conversation.ConversationNotExistsError:
             raise NotFound("Conversation Not Exists.")

api/controllers/service_api/app/file_preview.py

@@ -5,6 +5,7 @@ from flask import Response, request
 from flask_restx import Resource
 from pydantic import BaseModel, Field
 
+from controllers.common.file_response import enforce_download_for_html
 from controllers.common.schema import register_schema_model
 from controllers.service_api import service_api_ns
 from controllers.service_api.app.error import (
@@ -183,6 +184,13 @@ class FilePreviewApi(Resource):
             # Override content-type for downloads to force download
             response.headers["Content-Type"] = "application/octet-stream"
 
+        enforce_download_for_html(
+            response,
+            mime_type=upload_file.mime_type,
+            filename=upload_file.name,
+            extension=upload_file.extension,
+        )
+
         # Add caching headers for performance
         response.headers["Cache-Control"] = "public, max-age=3600"  # Cache for 1 hour
 

api/controllers/web/app.py

@@ -1,14 +1,13 @@
 import logging
 
 from flask import request
-from flask_restx import Resource, marshal_with, reqparse
+from flask_restx import Resource, marshal_with
+from pydantic import BaseModel, ConfigDict, Field
 from werkzeug.exceptions import Unauthorized
 
 from constants import HEADER_NAME_APP_CODE
 from controllers.common import fields
-from controllers.web import web_ns
-from controllers.web.error import AppUnavailableError
-from controllers.web.wraps import WebApiResource
+from controllers.common.schema import register_schema_models
 from core.app.app_config.common.parameters_mapping import get_parameters_from_feature_dict
 from libs.passport import PassportService
 from libs.token import extract_webapp_passport
@@ -18,9 +17,23 @@ from services.enterprise.enterprise_service import EnterpriseService
 from services.feature_service import FeatureService
 from services.webapp_auth_service import WebAppAuthService
 
+from . import web_ns
+from .error import AppUnavailableError
+from .wraps import WebApiResource
+
 logger = logging.getLogger(__name__)
 
 
+class AppAccessModeQuery(BaseModel):
+    model_config = ConfigDict(populate_by_name=True)
+
+    app_id: str | None = Field(default=None, alias="appId", description="Application ID")
+    app_code: str | None = Field(default=None, alias="appCode", description="Application code")
+
+
+register_schema_models(web_ns, AppAccessModeQuery)
+
+
 @web_ns.route("/parameters")
 class AppParameterApi(WebApiResource):
     """Resource for app variables."""
@@ -96,21 +109,16 @@ class AppAccessMode(Resource):
         }
     )
     def get(self):
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("appId", type=str, required=False, location="args")
-            .add_argument("appCode", type=str, required=False, location="args")
-        )
-        args = parser.parse_args()
+        raw_args = request.args.to_dict()
+        args = AppAccessModeQuery.model_validate(raw_args)
 
         features = FeatureService.get_system_features()
         if not features.webapp_auth.enabled:
             return {"accessMode": "public"}
 
-        app_id = args.get("appId")
-        if args.get("appCode"):
-            app_code = args["appCode"]
-            app_id = AppService.get_app_id_by_code(app_code)
+        app_id = args.app_id
+        if args.app_code:
+            app_id = AppService.get_app_id_by_code(args.app_code)
 
         if not app_id:
             raise ValueError("appId or appCode must be provided")

api/controllers/web/forgot_password.py

@@ -2,10 +2,12 @@ import base64
 import secrets
 
 from flask import request
-from flask_restx import Resource, reqparse
+from flask_restx import Resource
+from pydantic import BaseModel, Field, field_validator
 from sqlalchemy import select
 from sqlalchemy.orm import Session
 
+from controllers.common.schema import register_schema_models
 from controllers.console.auth.error import (
     AuthenticationFailedError,
     EmailCodeError,
@@ -18,14 +20,40 @@ from controllers.console.error import EmailSendIpLimitError
 from controllers.console.wraps import email_password_login_enabled, only_edition_enterprise, setup_required
 from controllers.web import web_ns
 from extensions.ext_database import db
-from libs.helper import email, extract_remote_ip
+from libs.helper import EmailStr, extract_remote_ip
 from libs.password import hash_password, valid_password
 from models import Account
 from services.account_service import AccountService
 
 
+class ForgotPasswordSendPayload(BaseModel):
+    email: EmailStr
+    language: str | None = None
+
+
+class ForgotPasswordCheckPayload(BaseModel):
+    email: EmailStr
+    code: str
+    token: str = Field(min_length=1)
+
+
+class ForgotPasswordResetPayload(BaseModel):
+    token: str = Field(min_length=1)
+    new_password: str
+    password_confirm: str
+
+    @field_validator("new_password", "password_confirm")
+    @classmethod
+    def validate_password(cls, value: str) -> str:
+        return valid_password(value)
+
+
+register_schema_models(web_ns, ForgotPasswordSendPayload, ForgotPasswordCheckPayload, ForgotPasswordResetPayload)
+
+
 @web_ns.route("/forgot-password")
 class ForgotPasswordSendEmailApi(Resource):
+    @web_ns.expect(web_ns.models[ForgotPasswordSendPayload.__name__])
     @only_edition_enterprise
     @setup_required
     @email_password_login_enabled
@@ -40,35 +68,31 @@ class ForgotPasswordSendEmailApi(Resource):
         }
     )
     def post(self):
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("email", type=email, required=True, location="json")
-            .add_argument("language", type=str, required=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = ForgotPasswordSendPayload.model_validate(web_ns.payload or {})
 
         ip_address = extract_remote_ip(request)
         if AccountService.is_email_send_ip_limit(ip_address):
             raise EmailSendIpLimitError()
 
-        if args["language"] is not None and args["language"] == "zh-Hans":
+        if payload.language == "zh-Hans":
             language = "zh-Hans"
         else:
             language = "en-US"
 
         with Session(db.engine) as session:
-            account = session.execute(select(Account).filter_by(email=args["email"])).scalar_one_or_none()
+            account = session.execute(select(Account).filter_by(email=payload.email)).scalar_one_or_none()
         token = None
         if account is None:
             raise AuthenticationFailedError()
         else:
-            token = AccountService.send_reset_password_email(account=account, email=args["email"], language=language)
+            token = AccountService.send_reset_password_email(account=account, email=payload.email, language=language)
 
         return {"result": "success", "data": token}
 
 
 @web_ns.route("/forgot-password/validity")
 class ForgotPasswordCheckApi(Resource):
+    @web_ns.expect(web_ns.models[ForgotPasswordCheckPayload.__name__])
     @only_edition_enterprise
     @setup_required
     @email_password_login_enabled
@@ -78,45 +102,40 @@ class ForgotPasswordCheckApi(Resource):
         responses={200: "Token is valid", 400: "Bad request - invalid token format", 401: "Invalid or expired token"}
     )
     def post(self):
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("email", type=str, required=True, location="json")
-            .add_argument("code", type=str, required=True, location="json")
-            .add_argument("token", type=str, required=True, nullable=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = ForgotPasswordCheckPayload.model_validate(web_ns.payload or {})
 
-        user_email = args["email"]
+        user_email = payload.email
 
-        is_forgot_password_error_rate_limit = AccountService.is_forgot_password_error_rate_limit(args["email"])
+        is_forgot_password_error_rate_limit = AccountService.is_forgot_password_error_rate_limit(payload.email)
         if is_forgot_password_error_rate_limit:
             raise EmailPasswordResetLimitError()
 
-        token_data = AccountService.get_reset_password_data(args["token"])
+        token_data = AccountService.get_reset_password_data(payload.token)
         if token_data is None:
             raise InvalidTokenError()
 
         if user_email != token_data.get("email"):
             raise InvalidEmailError()
 
-        if args["code"] != token_data.get("code"):
-            AccountService.add_forgot_password_error_rate_limit(args["email"])
+        if payload.code != token_data.get("code"):
+            AccountService.add_forgot_password_error_rate_limit(payload.email)
             raise EmailCodeError()
 
         # Verified, revoke the first token
-        AccountService.revoke_reset_password_token(args["token"])
+        AccountService.revoke_reset_password_token(payload.token)
 
         # Refresh token data by generating a new token
         _, new_token = AccountService.generate_reset_password_token(
-            user_email, code=args["code"], additional_data={"phase": "reset"}
+            user_email, code=payload.code, additional_data={"phase": "reset"}
         )
 
-        AccountService.reset_forgot_password_error_rate_limit(args["email"])
+        AccountService.reset_forgot_password_error_rate_limit(payload.email)
         return {"is_valid": True, "email": token_data.get("email"), "token": new_token}
 
 
 @web_ns.route("/forgot-password/resets")
 class ForgotPasswordResetApi(Resource):
+    @web_ns.expect(web_ns.models[ForgotPasswordResetPayload.__name__])
     @only_edition_enterprise
     @setup_required
     @email_password_login_enabled
@@ -131,20 +150,14 @@ class ForgotPasswordResetApi(Resource):
         }
     )
     def post(self):
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("token", type=str, required=True, nullable=False, location="json")
-            .add_argument("new_password", type=valid_password, required=True, nullable=False, location="json")
-            .add_argument("password_confirm", type=valid_password, required=True, nullable=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = ForgotPasswordResetPayload.model_validate(web_ns.payload or {})
 
         # Validate passwords match
-        if args["new_password"] != args["password_confirm"]:
+        if payload.new_password != payload.password_confirm:
             raise PasswordMismatchError()
 
         # Validate token and get reset data
-        reset_data = AccountService.get_reset_password_data(args["token"])
+        reset_data = AccountService.get_reset_password_data(payload.token)
         if not reset_data:
             raise InvalidTokenError()
         # Must use token in reset phase
@@ -152,11 +165,11 @@ class ForgotPasswordResetApi(Resource):
             raise InvalidTokenError()
 
         # Revoke token to prevent reuse
-        AccountService.revoke_reset_password_token(args["token"])
+        AccountService.revoke_reset_password_token(payload.token)
 
         # Generate secure salt and hash password
         salt = secrets.token_bytes(16)
-        password_hashed = hash_password(args["new_password"], salt)
+        password_hashed = hash_password(payload.new_password, salt)
 
         email = reset_data.get("email", "")
 
@@ -170,7 +183,7 @@ class ForgotPasswordResetApi(Resource):
 
         return {"result": "success"}
 
-    def _update_existing_account(self, account, password_hashed, salt, session):
+    def _update_existing_account(self, account: Account, password_hashed, salt, session):
         # Update existing account credentials
         account.password = base64.b64encode(password_hashed).decode()
         account.password_salt = base64.b64encode(salt).decode()

api/controllers/web/login.py

@@ -10,7 +10,12 @@ from controllers.console.auth.error import (
     InvalidEmailError,
 )
 from controllers.console.error import AccountBannedError
-from controllers.console.wraps import only_edition_enterprise, setup_required
+from controllers.console.wraps import (
+    decrypt_code_field,
+    decrypt_password_field,
+    only_edition_enterprise,
+    setup_required,
+)
 from controllers.web import web_ns
 from controllers.web.wraps import decode_jwt_token
 from libs.helper import email
@@ -42,6 +47,7 @@ class LoginApi(Resource):
             404: "Account not found",
         }
     )
+    @decrypt_password_field
     def post(self):
         """Authenticate user and login."""
         parser = (
@@ -181,6 +187,7 @@ class EmailCodeLoginApi(Resource):
             404: "Account not found",
         }
     )
+    @decrypt_code_field
     def post(self):
         parser = (
             reqparse.RequestParser()

api/controllers/web/message.py

@@ -1,9 +1,12 @@
 import logging
+from typing import Literal
 
-from flask_restx import fields, marshal_with, reqparse
-from flask_restx.inputs import int_range
+from flask import request
+from flask_restx import fields, marshal_with
+from pydantic import BaseModel, Field, field_validator
 from werkzeug.exceptions import InternalServerError, NotFound
 
+from controllers.common.schema import register_schema_models
 from controllers.web import web_ns
 from controllers.web.error import (
     AppMoreLikeThisDisabledError,
@@ -38,6 +41,33 @@ from services.message_service import MessageService
 logger = logging.getLogger(__name__)
 
 
+class MessageListQuery(BaseModel):
+    conversation_id: str = Field(description="Conversation UUID")
+    first_id: str | None = Field(default=None, description="First message ID for pagination")
+    limit: int = Field(default=20, ge=1, le=100, description="Number of messages to return (1-100)")
+
+    @field_validator("conversation_id", "first_id")
+    @classmethod
+    def validate_uuid(cls, value: str | None) -> str | None:
+        if value is None:
+            return value
+        return uuid_value(value)
+
+
+class MessageFeedbackPayload(BaseModel):
+    rating: Literal["like", "dislike"] | None = Field(default=None, description="Feedback rating")
+    content: str | None = Field(default=None, description="Feedback content")
+
+
+class MessageMoreLikeThisQuery(BaseModel):
+    response_mode: Literal["blocking", "streaming"] = Field(
+        description="Response mode",
+    )
+
+
+register_schema_models(web_ns, MessageListQuery, MessageFeedbackPayload, MessageMoreLikeThisQuery)
+
+
 @web_ns.route("/messages")
 class MessageListApi(WebApiResource):
     message_fields = {
@@ -68,7 +98,11 @@ class MessageListApi(WebApiResource):
     @web_ns.doc(
         params={
             "conversation_id": {"description": "Conversation UUID", "type": "string", "required": True},
-            "first_id": {"description": "First message ID for pagination", "type": "string", "required": False},
+            "first_id": {
+                "description": "First message ID for pagination",
+                "type": "string",
+                "required": False,
+            },
             "limit": {
                 "description": "Number of messages to return (1-100)",
                 "type": "integer",
@@ -93,17 +127,12 @@ class MessageListApi(WebApiResource):
         if app_mode not in {AppMode.CHAT, AppMode.AGENT_CHAT, AppMode.ADVANCED_CHAT}:
             raise NotChatAppError()
 
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("conversation_id", required=True, type=uuid_value, location="args")
-            .add_argument("first_id", type=uuid_value, location="args")
-            .add_argument("limit", type=int_range(1, 100), required=False, default=20, location="args")
-        )
-        args = parser.parse_args()
+        raw_args = request.args.to_dict()
+        query = MessageListQuery.model_validate(raw_args)
 
         try:
             return MessageService.pagination_by_first_id(
-                app_model, end_user, args["conversation_id"], args["first_id"], args["limit"]
+                app_model, end_user, query.conversation_id, query.first_id, query.limit
             )
         except ConversationNotExistsError:
             raise NotFound("Conversation Not Exists.")
@@ -128,7 +157,7 @@ class MessageFeedbackApi(WebApiResource):
                 "enum": ["like", "dislike"],
                 "required": False,
             },
-            "content": {"description": "Feedback content/comment", "type": "string", "required": False},
+            "content": {"description": "Feedback content", "type": "string", "required": False},
         }
     )
     @web_ns.doc(
@@ -145,20 +174,15 @@ class MessageFeedbackApi(WebApiResource):
     def post(self, app_model, end_user, message_id):
         message_id = str(message_id)
 
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("rating", type=str, choices=["like", "dislike", None], location="json")
-            .add_argument("content", type=str, location="json", default=None)
-        )
-        args = parser.parse_args()
+        payload = MessageFeedbackPayload.model_validate(web_ns.payload or {})
 
         try:
             MessageService.create_feedback(
                 app_model=app_model,
                 message_id=message_id,
                 user=end_user,
-                rating=args.get("rating"),
-                content=args.get("content"),
+                rating=payload.rating,
+                content=payload.content,
             )
         except MessageNotExistsError:
             raise NotFound("Message Not Exists.")
@@ -170,17 +194,7 @@ class MessageFeedbackApi(WebApiResource):
 class MessageMoreLikeThisApi(WebApiResource):
     @web_ns.doc("Generate More Like This")
     @web_ns.doc(description="Generate a new completion similar to an existing message (completion apps only).")
-    @web_ns.doc(
-        params={
-            "message_id": {"description": "Message UUID", "type": "string", "required": True},
-            "response_mode": {
-                "description": "Response mode",
-                "type": "string",
-                "enum": ["blocking", "streaming"],
-                "required": True,
-            },
-        }
-    )
+    @web_ns.expect(web_ns.models[MessageMoreLikeThisQuery.__name__])
     @web_ns.doc(
         responses={
             200: "Success",
@@ -197,12 +211,10 @@ class MessageMoreLikeThisApi(WebApiResource):
 
         message_id = str(message_id)
 
-        parser = reqparse.RequestParser().add_argument(
-            "response_mode", type=str, required=True, choices=["blocking", "streaming"], location="args"
-        )
-        args = parser.parse_args()
+        raw_args = request.args.to_dict()
+        query = MessageMoreLikeThisQuery.model_validate(raw_args)
 
-        streaming = args["response_mode"] == "streaming"
+        streaming = query.response_mode == "streaming"
 
         try:
             response = AppGenerateService.generate_more_like_this(

api/controllers/web/remote_files.py

@@ -1,7 +1,8 @@
 import urllib.parse
 
 import httpx
-from flask_restx import marshal_with, reqparse
+from flask_restx import marshal_with
+from pydantic import BaseModel, Field, HttpUrl
 
 import services
 from controllers.common import helpers
@@ -10,14 +11,23 @@ from controllers.common.errors import (
     RemoteFileUploadError,
     UnsupportedFileTypeError,
 )
-from controllers.web import web_ns
-from controllers.web.wraps import WebApiResource
 from core.file import helpers as file_helpers
 from core.helper import ssrf_proxy
 from extensions.ext_database import db
 from fields.file_fields import build_file_with_signed_url_model, build_remote_file_info_model
 from services.file_service import FileService
 
+from ..common.schema import register_schema_models
+from . import web_ns
+from .wraps import WebApiResource
+
+
+class RemoteFileUploadPayload(BaseModel):
+    url: HttpUrl = Field(description="Remote file URL")
+
+
+register_schema_models(web_ns, RemoteFileUploadPayload)
+
 
 @web_ns.route("/remote-files/<path:url>")
 class RemoteFileInfoApi(WebApiResource):
@@ -97,10 +107,8 @@ class RemoteFileUploadApi(WebApiResource):
             FileTooLargeError: File exceeds size limit
             UnsupportedFileTypeError: File type not supported
         """
-        parser = reqparse.RequestParser().add_argument("url", type=str, required=True, help="URL is required")
-        args = parser.parse_args()
-
-        url = args["url"]
+        payload = RemoteFileUploadPayload.model_validate(web_ns.payload or {})
+        url = str(payload.url)
 
         try:
             resp = ssrf_proxy.head(url=url)

api/core/app/apps/base_app_generator.py

@@ -105,8 +105,9 @@ class BaseAppGenerator:
             variable_entity.type in {VariableEntityType.FILE, VariableEntityType.FILE_LIST}
             and not variable_entity.required
         ):
-            # Treat empty string (frontend default) or empty list as unset
-            if not value and isinstance(value, (str, list)):
+            # Treat empty string (frontend default) as unset
+            # For FILE_LIST, allow empty list [] to pass through
+            if isinstance(value, str) and not value:
                 return None
 
         if variable_entity.type in {

api/core/helper/ssrf_proxy.py

@@ -72,6 +72,22 @@ def _get_ssrf_client(ssl_verify_enabled: bool) -> httpx.Client:
     )
 
 
+def _get_user_provided_host_header(headers: dict | None) -> str | None:
+    """
+    Extract the user-provided Host header from the headers dict.
+
+    This is needed because when using a forward proxy, httpx may override the Host header.
+    We preserve the user's explicit Host header to support virtual hosting and other use cases.
+    """
+    if not headers:
+        return None
+    # Case-insensitive lookup for Host header
+    for key, value in headers.items():
+        if key.lower() == "host":
+            return value
+    return None
+
+
 def make_request(method, url, max_retries=SSRF_DEFAULT_MAX_RETRIES, **kwargs):
     if "allow_redirects" in kwargs:
         allow_redirects = kwargs.pop("allow_redirects")
@@ -90,10 +106,24 @@ def make_request(method, url, max_retries=SSRF_DEFAULT_MAX_RETRIES, **kwargs):
     verify_option = kwargs.pop("ssl_verify", dify_config.HTTP_REQUEST_NODE_SSL_VERIFY)
     client = _get_ssrf_client(verify_option)
 
+    # Preserve user-provided Host header
+    # When using a forward proxy, httpx may override the Host header based on the URL.
+    # We extract and preserve any explicitly set Host header to support virtual hosting.
+    headers = kwargs.get("headers", {})
+    user_provided_host = _get_user_provided_host_header(headers)
+
     retries = 0
     while retries <= max_retries:
         try:
+            # Build the request manually to preserve the Host header
+            # httpx may override the Host header when using a proxy, so we use
+            # the request API to explicitly set headers before sending
+            headers = {k: v for k, v in headers.items() if k.lower() != "host"}
+            if user_provided_host is not None:
+                headers["host"] = user_provided_host
+            kwargs["headers"] = headers
             response = client.request(method=method, url=url, **kwargs)
+
             # Check for SSRF protection by Squid proxy
             if response.status_code in (401, 403):
                 # Check if this is a Squid SSRF rejection

api/core/indexing_runner.py

@@ -396,7 +396,7 @@ class IndexingRunner:
                 datasource_type=DatasourceType.NOTION,
                 notion_info=NotionInfo.model_validate(
                     {
-                        "credential_id": data_source_info["credential_id"],
+                        "credential_id": data_source_info.get("credential_id"),
                         "notion_workspace_id": data_source_info["notion_workspace_id"],
                         "notion_obj_id": data_source_info["notion_page_id"],
                         "notion_page_type": data_source_info["type"],

api/core/mcp/auth/auth_flow.py

@@ -47,7 +47,11 @@ def build_protected_resource_metadata_discovery_urls(
     """
     Build a list of URLs to try for Protected Resource Metadata discovery.
 
-    Per SEP-985, supports fallback when discovery fails at one URL.
+    Per RFC 9728 Section 5.1, supports fallback when discovery fails at one URL.
+    Priority order:
+    1. URL from WWW-Authenticate header (if provided)
+    2. Well-known URI with path: https://example.com/.well-known/oauth-protected-resource/public/mcp
+    3. Well-known URI at root: https://example.com/.well-known/oauth-protected-resource
     """
     urls = []
 
@@ -58,9 +62,18 @@ def build_protected_resource_metadata_discovery_urls(
     # Fallback: construct from server URL
     parsed = urlparse(server_url)
     base_url = f"{parsed.scheme}://{parsed.netloc}"
-    fallback_url = urljoin(base_url, "/.well-known/oauth-protected-resource")
-    if fallback_url not in urls:
-        urls.append(fallback_url)
+    path = parsed.path.rstrip("/")
+
+    # Priority 2: With path insertion (e.g., /.well-known/oauth-protected-resource/public/mcp)
+    if path:
+        path_url = f"{base_url}/.well-known/oauth-protected-resource{path}"
+        if path_url not in urls:
+            urls.append(path_url)
+
+    # Priority 3: At root (e.g., /.well-known/oauth-protected-resource)
+    root_url = f"{base_url}/.well-known/oauth-protected-resource"
+    if root_url not in urls:
+        urls.append(root_url)
 
     return urls
 
@@ -71,30 +84,34 @@ def build_oauth_authorization_server_metadata_discovery_urls(auth_server_url: st
 
     Supports both OAuth 2.0 (RFC 8414) and OpenID Connect discovery.
 
-    Per RFC 8414 section 3:
-    - If issuer has no path: https://example.com/.well-known/oauth-authorization-server
-    - If issuer has path: https://example.com/.well-known/oauth-authorization-server{path}
-
-    Example:
-    - issuer: https://example.com/oauth
-    - metadata: https://example.com/.well-known/oauth-authorization-server/oauth
+    Per RFC 8414 section 3.1 and section 5, try all possible endpoints:
+    - OAuth 2.0 with path insertion: https://example.com/.well-known/oauth-authorization-server/tenant1
+    - OpenID Connect with path insertion: https://example.com/.well-known/openid-configuration/tenant1
+    - OpenID Connect path appending: https://example.com/tenant1/.well-known/openid-configuration
+    - OAuth 2.0 at root: https://example.com/.well-known/oauth-authorization-server
+    - OpenID Connect at root: https://example.com/.well-known/openid-configuration
     """
     urls = []
     base_url = auth_server_url or server_url
 
     parsed = urlparse(base_url)
     base = f"{parsed.scheme}://{parsed.netloc}"
-    path = parsed.path.rstrip("/")  # Remove trailing slash
+    path = parsed.path.rstrip("/")
+    # OAuth 2.0 Authorization Server Metadata at root (MCP-03-26)
+    urls.append(f"{base}/.well-known/oauth-authorization-server")
 
-    # Try OpenID Connect discovery first (more common)
-    urls.append(urljoin(base + "/", ".well-known/openid-configuration"))
+    # OpenID Connect Discovery at root
+    urls.append(f"{base}/.well-known/openid-configuration")
 
-    # OAuth 2.0 Authorization Server Metadata (RFC 8414)
-    # Include the path component if present in the issuer URL
     if path:
-        urls.append(urljoin(base, f".well-known/oauth-authorization-server{path}"))
-    else:
-        urls.append(urljoin(base, ".well-known/oauth-authorization-server"))
+        # OpenID Connect Discovery with path insertion
+        urls.append(f"{base}/.well-known/openid-configuration{path}")
+
+        # OpenID Connect Discovery path appending
+        urls.append(f"{base}{path}/.well-known/openid-configuration")
+
+        # OAuth 2.0 Authorization Server Metadata with path insertion
+        urls.append(f"{base}/.well-known/oauth-authorization-server{path}")
 
     return urls
 

api/core/mcp/client/sse_client.py

@@ -61,6 +61,7 @@ class SSETransport:
         self.timeout = timeout
         self.sse_read_timeout = sse_read_timeout
         self.endpoint_url: str | None = None
+        self.event_source: EventSource | None = None
 
     def _validate_endpoint_url(self, endpoint_url: str) -> bool:
         """Validate that the endpoint URL matches the connection origin.
@@ -237,6 +238,9 @@ class SSETransport:
         write_queue: WriteQueue = queue.Queue()
         status_queue: StatusQueue = queue.Queue()
 
+        # Store event_source for graceful shutdown
+        self.event_source = event_source
+
         # Start SSE reader thread
         executor.submit(self.sse_reader, event_source, read_queue, status_queue)
 
@@ -296,6 +300,13 @@ def sse_client(
         logger.exception("Error connecting to SSE endpoint")
         raise
     finally:
+        # Close the SSE connection to unblock the reader thread
+        if transport.event_source is not None:
+            try:
+                transport.event_source.response.close()
+            except RuntimeError:
+                pass
+
         # Clean up queues
         if read_queue:
             read_queue.put(None)

api/core/mcp/client/streamable_client.py

@@ -8,6 +8,7 @@ and session management.
 
 import logging
 import queue
+import threading
 from collections.abc import Callable, Generator
 from concurrent.futures import ThreadPoolExecutor
 from contextlib import contextmanager
@@ -103,6 +104,9 @@ class StreamableHTTPTransport:
             CONTENT_TYPE: JSON,
             **self.headers,
         }
+        self.stop_event = threading.Event()
+        self._active_responses: list[httpx.Response] = []
+        self._lock = threading.Lock()
 
     def _update_headers_with_session(self, base_headers: dict[str, str]) -> dict[str, str]:
         """Update headers with session ID if available."""
@@ -111,6 +115,30 @@ class StreamableHTTPTransport:
             headers[MCP_SESSION_ID] = self.session_id
         return headers
 
+    def _register_response(self, response: httpx.Response):
+        """Register a response for cleanup on shutdown."""
+        with self._lock:
+            self._active_responses.append(response)
+
+    def _unregister_response(self, response: httpx.Response):
+        """Unregister a response after it's closed."""
+        with self._lock:
+            try:
+                self._active_responses.remove(response)
+            except ValueError as e:
+                logger.debug("Ignoring error during response unregister: %s", e)
+
+    def close_active_responses(self):
+        """Close all active SSE connections to unblock threads."""
+        with self._lock:
+            responses_to_close = list(self._active_responses)
+            self._active_responses.clear()
+            for response in responses_to_close:
+                try:
+                    response.close()
+                except RuntimeError as e:
+                    logger.debug("Ignoring error during active response close: %s", e)
+
     def _is_initialization_request(self, message: JSONRPCMessage) -> bool:
         """Check if the message is an initialization request."""
         return isinstance(message.root, JSONRPCRequest) and message.root.method == "initialize"
@@ -195,11 +223,21 @@ class StreamableHTTPTransport:
                 event_source.response.raise_for_status()
                 logger.debug("GET SSE connection established")
 
-                for sse in event_source.iter_sse():
-                    self._handle_sse_event(sse, server_to_client_queue)
+                # Register response for cleanup
+                self._register_response(event_source.response)
+
+                try:
+                    for sse in event_source.iter_sse():
+                        if self.stop_event.is_set():
+                            logger.debug("GET stream received stop signal")
+                            break
+                        self._handle_sse_event(sse, server_to_client_queue)
+                finally:
+                    self._unregister_response(event_source.response)
 
         except Exception as exc:
-            logger.debug("GET stream error (non-fatal): %s", exc)
+            if not self.stop_event.is_set():
+                logger.debug("GET stream error (non-fatal): %s", exc)
 
     def _handle_resumption_request(self, ctx: RequestContext):
         """Handle a resumption request using GET with SSE."""
@@ -224,15 +262,24 @@ class StreamableHTTPTransport:
             event_source.response.raise_for_status()
             logger.debug("Resumption GET SSE connection established")
 
-            for sse in event_source.iter_sse():
-                is_complete = self._handle_sse_event(
-                    sse,
-                    ctx.server_to_client_queue,
-                    original_request_id,
-                    ctx.metadata.on_resumption_token_update if ctx.metadata else None,
-                )
-                if is_complete:
-                    break
+            # Register response for cleanup
+            self._register_response(event_source.response)
+
+            try:
+                for sse in event_source.iter_sse():
+                    if self.stop_event.is_set():
+                        logger.debug("Resumption stream received stop signal")
+                        break
+                    is_complete = self._handle_sse_event(
+                        sse,
+                        ctx.server_to_client_queue,
+                        original_request_id,
+                        ctx.metadata.on_resumption_token_update if ctx.metadata else None,
+                    )
+                    if is_complete:
+                        break
+            finally:
+                self._unregister_response(event_source.response)
 
     def _handle_post_request(self, ctx: RequestContext):
         """Handle a POST request with response processing."""
@@ -295,17 +342,27 @@ class StreamableHTTPTransport:
     def _handle_sse_response(self, response: httpx.Response, ctx: RequestContext):
         """Handle SSE response from the server."""
         try:
+            # Register response for cleanup
+            self._register_response(response)
+
             event_source = EventSource(response)
-            for sse in event_source.iter_sse():
-                is_complete = self._handle_sse_event(
-                    sse,
-                    ctx.server_to_client_queue,
-                    resumption_callback=(ctx.metadata.on_resumption_token_update if ctx.metadata else None),
-                )
-                if is_complete:
-                    break
+            try:
+                for sse in event_source.iter_sse():
+                    if self.stop_event.is_set():
+                        logger.debug("SSE response stream received stop signal")
+                        break
+                    is_complete = self._handle_sse_event(
+                        sse,
+                        ctx.server_to_client_queue,
+                        resumption_callback=(ctx.metadata.on_resumption_token_update if ctx.metadata else None),
+                    )
+                    if is_complete:
+                        break
+            finally:
+                self._unregister_response(response)
         except Exception as e:
-            ctx.server_to_client_queue.put(e)
+            if not self.stop_event.is_set():
+                ctx.server_to_client_queue.put(e)
 
     def _handle_unexpected_content_type(
         self,
@@ -345,6 +402,11 @@ class StreamableHTTPTransport:
         """
         while True:
             try:
+                # Check if we should stop
+                if self.stop_event.is_set():
+                    logger.debug("Post writer received stop signal")
+                    break
+
                 # Read message from client queue with timeout to check stop_event periodically
                 session_message = client_to_server_queue.get(timeout=DEFAULT_QUEUE_READ_TIMEOUT)
                 if session_message is None:
@@ -381,7 +443,8 @@ class StreamableHTTPTransport:
             except queue.Empty:
                 continue
             except Exception as exc:
-                server_to_client_queue.put(exc)
+                if not self.stop_event.is_set():
+                    server_to_client_queue.put(exc)
 
     def terminate_session(self, client: httpx.Client):
         """Terminate the session by sending a DELETE request."""
@@ -465,6 +528,12 @@ def streamablehttp_client(
                     transport.get_session_id,
                 )
             finally:
+                # Set stop event to signal all threads to stop
+                transport.stop_event.set()
+
+                # Close all active SSE connections to unblock threads
+                transport.close_active_responses()
+
                 if transport.session_id and terminate_on_close:
                     transport.terminate_session(client)
 

api/core/mcp/mcp_client.py

@@ -59,7 +59,7 @@ class MCPClient:
             try:
                 logger.debug("Not supported method %s found in URL path, trying default 'mcp' method.", method_name)
                 self.connect_server(sse_client, "sse")
-            except MCPConnectionError:
+            except (MCPConnectionError, ValueError):
                 logger.debug("MCP connection failed with 'sse', falling back to 'mcp' method.")
                 self.connect_server(streamablehttp_client, "mcp")
 

api/core/ops/utils.py

@@ -54,7 +54,7 @@ def generate_dotted_order(run_id: str, start_time: Union[str, datetime], parent_
     generate dotted_order for langsmith
     """
     start_time = datetime.fromisoformat(start_time) if isinstance(start_time, str) else start_time
-    timestamp = start_time.strftime("%Y%m%dT%H%M%S%f")[:-3] + "Z"
+    timestamp = start_time.strftime("%Y%m%dT%H%M%S%f") + "Z"
     current_segment = f"{timestamp}{run_id}"
 
     if parent_dotted_order is None:

api/core/rag/datasource/keyword/jieba/jieba.py

@@ -90,13 +90,17 @@ class Jieba(BaseKeyword):
         sorted_chunk_indices = self._retrieve_ids_by_query(keyword_table or {}, query, k)
 
         documents = []
+
+        segment_query_stmt = db.session.query(DocumentSegment).where(
+            DocumentSegment.dataset_id == self.dataset.id, DocumentSegment.index_node_id.in_(sorted_chunk_indices)
+        )
+        if document_ids_filter:
+            segment_query_stmt = segment_query_stmt.where(DocumentSegment.document_id.in_(document_ids_filter))
+
+        segments = db.session.execute(segment_query_stmt).scalars().all()
+        segment_map = {segment.index_node_id: segment for segment in segments}
         for chunk_index in sorted_chunk_indices:
-            segment_query = db.session.query(DocumentSegment).where(
-                DocumentSegment.dataset_id == self.dataset.id, DocumentSegment.index_node_id == chunk_index
-            )
-            if document_ids_filter:
-                segment_query = segment_query.where(DocumentSegment.document_id.in_(document_ids_filter))
-            segment = segment_query.first()
+            segment = segment_map.get(chunk_index)
 
             if segment:
                 documents.append(

api/core/rag/datasource/retrieval_service.py

@@ -7,6 +7,7 @@ from sqlalchemy import select
 from sqlalchemy.orm import Session, load_only
 
 from configs import dify_config
+from core.db.session_factory import session_factory
 from core.model_manager import ModelManager
 from core.model_runtime.entities.model_entities import ModelType
 from core.rag.data_post_processor.data_post_processor import DataPostProcessor
@@ -138,37 +139,47 @@ class RetrievalService:
 
     @classmethod
     def _deduplicate_documents(cls, documents: list[Document]) -> list[Document]:
-        """Deduplicate documents based on doc_id to avoid duplicate chunks in hybrid search."""
+        """Deduplicate documents in O(n) while preserving first-seen order.
+
+        Rules:
+        - For provider == "dify" and metadata["doc_id"] exists: keep the doc with the highest
+          metadata["score"] among duplicates; if a later duplicate has no score, ignore it.
+        - For non-dify documents (or dify without doc_id): deduplicate by content key
+          (provider, page_content), keeping the first occurrence.
+        """
         if not documents:
             return documents
 
-        unique_documents = []
-        seen_doc_ids = set()
+        # Map of dedup key -> chosen Document
+        chosen: dict[tuple, Document] = {}
+        # Preserve the order of first appearance of each dedup key
+        order: list[tuple] = []
 
-        for document in documents:
-            # For dify provider documents, use doc_id for deduplication
-            if document.provider == "dify" and document.metadata is not None and "doc_id" in document.metadata:
-                doc_id = document.metadata["doc_id"]
-                if doc_id not in seen_doc_ids:
-                    seen_doc_ids.add(doc_id)
-                    unique_documents.append(document)
-                # If duplicate, keep the one with higher score
-                elif "score" in document.metadata:
-                    # Find existing document with same doc_id and compare scores
-                    for i, existing_doc in enumerate(unique_documents):
-                        if (
-                            existing_doc.metadata
-                            and existing_doc.metadata.get("doc_id") == doc_id
-                            and existing_doc.metadata.get("score", 0) < document.metadata.get("score", 0)
-                        ):
-                            unique_documents[i] = document
-                            break
+        for doc in documents:
+            is_dify = doc.provider == "dify"
+            doc_id = (doc.metadata or {}).get("doc_id") if is_dify else None
+
+            if is_dify and doc_id:
+                key = ("dify", doc_id)
+                if key not in chosen:
+                    chosen[key] = doc
+                    order.append(key)
+                else:
+                    # Only replace if the new one has a score and it's strictly higher
+                    if "score" in doc.metadata:
+                        new_score = float(doc.metadata.get("score", 0.0))
+                        old_score = float(chosen[key].metadata.get("score", 0.0)) if chosen[key].metadata else 0.0
+                        if new_score > old_score:
+                            chosen[key] = doc
             else:
-                # For non-dify documents, use content-based deduplication
-                if document not in unique_documents:
-                    unique_documents.append(document)
+                # Content-based dedup for non-dify or dify without doc_id
+                content_key = (doc.provider or "dify", doc.page_content)
+                if content_key not in chosen:
+                    chosen[content_key] = doc
+                    order.append(content_key)
+                # If duplicate content appears, we keep the first occurrence (no score comparison)
 
-        return unique_documents
+        return [chosen[k] for k in order]
 
     @classmethod
     def _get_dataset(cls, dataset_id: str) -> Dataset | None:
@@ -370,161 +381,155 @@ class RetrievalService:
             records = []
             include_segment_ids = set()
             segment_child_map = {}
-            segment_file_map = {}
-            with Session(bind=db.engine, expire_on_commit=False) as session:
-                # Process documents
-                for document in documents:
-                    segment_id = None
-                    attachment_info = None
-                    child_chunk = None
-                    document_id = document.metadata.get("document_id")
-                    if document_id not in dataset_documents:
-                        continue
 
-                    dataset_document = dataset_documents[document_id]
-                    if not dataset_document:
-                        continue
+            valid_dataset_documents = {}
+            image_doc_ids: list[Any] = []
+            child_index_node_ids = []
+            index_node_ids = []
+            doc_to_document_map = {}
+            for document in documents:
+                document_id = document.metadata.get("document_id")
+                if document_id not in dataset_documents:
+                    continue
 
-                    if dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
-                        # Handle parent-child documents
-                        if document.metadata.get("doc_type") == DocType.IMAGE:
-                            attachment_info_dict = cls.get_segment_attachment_info(
-                                dataset_document.dataset_id,
-                                dataset_document.tenant_id,
-                                document.metadata.get("doc_id") or "",
-                                session,
-                            )
-                            if attachment_info_dict:
-                                attachment_info = attachment_info_dict["attachment_info"]
-                                segment_id = attachment_info_dict["segment_id"]
-                        else:
-                            child_index_node_id = document.metadata.get("doc_id")
-                            child_chunk_stmt = select(ChildChunk).where(ChildChunk.index_node_id == child_index_node_id)
-                            child_chunk = session.scalar(child_chunk_stmt)
+                dataset_document = dataset_documents[document_id]
+                if not dataset_document:
+                    continue
+                valid_dataset_documents[document_id] = dataset_document
 
-                            if not child_chunk:
-                                continue
-                            segment_id = child_chunk.segment_id
-
-                        if not segment_id:
-                            continue
-
-                        segment = (
-                            session.query(DocumentSegment)
-                            .where(
-                                DocumentSegment.dataset_id == dataset_document.dataset_id,
-                                DocumentSegment.enabled == True,
-                                DocumentSegment.status == "completed",
-                                DocumentSegment.id == segment_id,
-                            )
-                            .first()
-                        )
-
-                        if not segment:
-                            continue
-
-                        if segment.id not in include_segment_ids:
-                            include_segment_ids.add(segment.id)
-                            if child_chunk:
-                                child_chunk_detail = {
-                                    "id": child_chunk.id,
-                                    "content": child_chunk.content,
-                                    "position": child_chunk.position,
-                                    "score": document.metadata.get("score", 0.0),
-                                }
-                                map_detail = {
-                                    "max_score": document.metadata.get("score", 0.0),
-                                    "child_chunks": [child_chunk_detail],
-                                }
-                                segment_child_map[segment.id] = map_detail
-                            record = {
-                                "segment": segment,
-                            }
-                            if attachment_info:
-                                segment_file_map[segment.id] = [attachment_info]
-                            records.append(record)
-                        else:
-                            if child_chunk:
-                                child_chunk_detail = {
-                                    "id": child_chunk.id,
-                                    "content": child_chunk.content,
-                                    "position": child_chunk.position,
-                                    "score": document.metadata.get("score", 0.0),
-                                }
-                                if segment.id in segment_child_map:
-                                    segment_child_map[segment.id]["child_chunks"].append(child_chunk_detail)
-                                    segment_child_map[segment.id]["max_score"] = max(
-                                        segment_child_map[segment.id]["max_score"], document.metadata.get("score", 0.0)
-                                    )
-                                else:
-                                    segment_child_map[segment.id] = {
-                                        "max_score": document.metadata.get("score", 0.0),
-                                        "child_chunks": [child_chunk_detail],
-                                    }
-                            if attachment_info:
-                                if segment.id in segment_file_map:
-                                    segment_file_map[segment.id].append(attachment_info)
-                                else:
-                                    segment_file_map[segment.id] = [attachment_info]
+                if dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
+                    doc_id = document.metadata.get("doc_id") or ""
+                    doc_to_document_map[doc_id] = document
+                    if document.metadata.get("doc_type") == DocType.IMAGE:
+                        image_doc_ids.append(doc_id)
                     else:
-                        # Handle normal documents
-                        segment = None
-                        if document.metadata.get("doc_type") == DocType.IMAGE:
-                            attachment_info_dict = cls.get_segment_attachment_info(
-                                dataset_document.dataset_id,
-                                dataset_document.tenant_id,
-                                document.metadata.get("doc_id") or "",
-                                session,
-                            )
-                            if attachment_info_dict:
-                                attachment_info = attachment_info_dict["attachment_info"]
-                                segment_id = attachment_info_dict["segment_id"]
-                                document_segment_stmt = select(DocumentSegment).where(
-                                    DocumentSegment.dataset_id == dataset_document.dataset_id,
-                                    DocumentSegment.enabled == True,
-                                    DocumentSegment.status == "completed",
-                                    DocumentSegment.id == segment_id,
-                                )
-                                segment = session.scalar(document_segment_stmt)
-                                if segment:
-                                    segment_file_map[segment.id] = [attachment_info]
-                        else:
-                            index_node_id = document.metadata.get("doc_id")
-                            if not index_node_id:
-                                continue
-                            document_segment_stmt = select(DocumentSegment).where(
-                                DocumentSegment.dataset_id == dataset_document.dataset_id,
-                                DocumentSegment.enabled == True,
-                                DocumentSegment.status == "completed",
-                                DocumentSegment.index_node_id == index_node_id,
-                            )
-                            segment = session.scalar(document_segment_stmt)
+                        child_index_node_ids.append(doc_id)
+                else:
+                    doc_id = document.metadata.get("doc_id") or ""
+                    doc_to_document_map[doc_id] = document
+                    if document.metadata.get("doc_type") == DocType.IMAGE:
+                        image_doc_ids.append(doc_id)
+                    else:
+                        index_node_ids.append(doc_id)
 
-                        if not segment:
-                            continue
-                        if segment.id not in include_segment_ids:
-                            include_segment_ids.add(segment.id)
-                            record = {
-                                "segment": segment,
-                                "score": document.metadata.get("score"),  # type: ignore
+            image_doc_ids = [i for i in image_doc_ids if i]
+            child_index_node_ids = [i for i in child_index_node_ids if i]
+            index_node_ids = [i for i in index_node_ids if i]
+
+            segment_ids = []
+            index_node_segments: list[DocumentSegment] = []
+            segments: list[DocumentSegment] = []
+            attachment_map = {}
+            child_chunk_map: dict[Any, Any] = {}
+            doc_segment_map = {}
+
+            with session_factory.create_session() as session:
+                attachments = cls.get_segment_attachment_infos(image_doc_ids, session)
+
+                for attachment in attachments:
+                    segment_ids.append(attachment["segment_id"])
+                    if attachment["segment_id"] in attachment_map:
+                        attachment_map[attachment["segment_id"]].append(attachment["attachment_info"])
+                    else:
+                        attachment_map[attachment["segment_id"]] = [attachment["attachment_info"]]
+                    if attachment["attachment_id"] in doc_segment_map:
+                        doc_segment_map[attachment["segment_id"]].append(attachment["attachment_id"])
+                    else:
+                        doc_segment_map[attachment["segment_id"]] = [attachment["attachment_id"]]
+                child_chunk_stmt = select(ChildChunk).where(ChildChunk.index_node_id.in_(child_index_node_ids))
+                child_index_nodes = session.execute(child_chunk_stmt).scalars().all()
+
+                for i in child_index_nodes:
+                    segment_ids.append(i.segment_id)
+                    if i.segment_id in child_chunk_map:
+                        child_chunk_map[i.segment_id].append(i)
+                    else:
+                        child_chunk_map[i.segment_id] = [i]
+                    if i.segment_id in doc_segment_map:
+                        doc_segment_map[i.segment_id].append(i.index_node_id)
+                    else:
+                        doc_segment_map[i.segment_id] = [i.index_node_id]
+
+                if index_node_ids:
+                    document_segment_stmt = select(DocumentSegment).where(
+                        DocumentSegment.enabled == True,
+                        DocumentSegment.status == "completed",
+                        DocumentSegment.index_node_id.in_(index_node_ids),
+                    )
+                    index_node_segments = session.execute(document_segment_stmt).scalars().all()  # type: ignore
+                    for index_node_segment in index_node_segments:
+                        doc_segment_map[index_node_segment.id] = [index_node_segment.index_node_id]
+                if segment_ids:
+                    document_segment_stmt = select(DocumentSegment).where(
+                        DocumentSegment.enabled == True,
+                        DocumentSegment.status == "completed",
+                        DocumentSegment.id.in_(segment_ids),
+                    )
+                    segments = session.execute(document_segment_stmt).scalars().all()  # type: ignore
+
+                if index_node_segments:
+                    segments.extend(index_node_segments)
+
+            for segment in segments:
+                child_chunks: list[ChildChunk] = child_chunk_map.get(segment.id, [])
+                attachment_infos: list[dict[str, Any]] = attachment_map.get(segment.id, [])
+                ds_dataset_document: DatasetDocument | None = valid_dataset_documents.get(segment.document_id)
+
+                if ds_dataset_document and ds_dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
+                    if segment.id not in include_segment_ids:
+                        include_segment_ids.add(segment.id)
+                        if child_chunks or attachment_infos:
+                            child_chunk_details = []
+                            max_score = 0.0
+                            for child_chunk in child_chunks:
+                                document = doc_to_document_map[child_chunk.index_node_id]
+                                child_chunk_detail = {
+                                    "id": child_chunk.id,
+                                    "content": child_chunk.content,
+                                    "position": child_chunk.position,
+                                    "score": document.metadata.get("score", 0.0) if document else 0.0,
+                                }
+                                child_chunk_details.append(child_chunk_detail)
+                                max_score = max(max_score, document.metadata.get("score", 0.0) if document else 0.0)
+                            for attachment_info in attachment_infos:
+                                file_document = doc_to_document_map[attachment_info["id"]]
+                                max_score = max(
+                                    max_score, file_document.metadata.get("score", 0.0) if file_document else 0.0
+                                )
+
+                            map_detail = {
+                                "max_score": max_score,
+                                "child_chunks": child_chunk_details,
                             }
-                            if attachment_info:
-                                segment_file_map[segment.id] = [attachment_info]
-                            records.append(record)
-                        else:
-                            if attachment_info:
-                                attachment_infos = segment_file_map.get(segment.id, [])
-                                if attachment_info not in attachment_infos:
-                                    attachment_infos.append(attachment_info)
-                                segment_file_map[segment.id] = attachment_infos
+                            segment_child_map[segment.id] = map_detail
+                        record = {
+                            "segment": segment,
+                        }
+                        records.append(record)
+                else:
+                    if segment.id not in include_segment_ids:
+                        include_segment_ids.add(segment.id)
+                        max_score = 0.0
+                        document = doc_to_document_map.get(segment.index_node_id)
+                        if document:
+                            max_score = max(max_score, document.metadata.get("score", 0.0))
+                        for attachment_info in attachment_infos:
+                            file_document = doc_to_document_map.get(attachment_info["id"])
+                            if file_document:
+                                max_score = max(max_score, file_document.metadata.get("score", 0.0))
+                        record = {
+                            "segment": segment,
+                            "score": max_score,
+                        }
+                        records.append(record)
 
             # Add child chunks information to records
             for record in records:
                 if record["segment"].id in segment_child_map:
                     record["child_chunks"] = segment_child_map[record["segment"].id].get("child_chunks")  # type: ignore
-                    record["score"] = segment_child_map[record["segment"].id]["max_score"]
-                if record["segment"].id in segment_file_map:
-                    record["files"] = segment_file_map[record["segment"].id]  # type: ignore[assignment]
+                    record["score"] = segment_child_map[record["segment"].id]["max_score"]  # type: ignore
+                if record["segment"].id in attachment_map:
+                    record["files"] = attachment_map[record["segment"].id]  # type: ignore[assignment]
 
             result = []
             for record in records:
@@ -536,6 +541,9 @@ class RetrievalService:
                 if not isinstance(child_chunks, list):
                     child_chunks = None
 
+                if child_chunks:
+                    child_chunks = sorted(child_chunks, key=lambda x: x.get("score", 0.0), reverse=True)
+
                 # Extract files, ensuring it's a list or None
                 files = record.get("files")
                 if not isinstance(files, list):
@@ -555,7 +563,7 @@ class RetrievalService:
                 )
                 result.append(retrieval_segment)
 
-            return result
+            return sorted(result, key=lambda x: x.score, reverse=True)
         except Exception as e:
             db.session.rollback()
             raise e
@@ -565,6 +573,8 @@ class RetrievalService:
         flask_app: Flask,
         retrieval_method: RetrievalMethod,
         dataset: Dataset,
+        all_documents: list[Document],
+        exceptions: list[str],
         query: str | None = None,
         top_k: int = 4,
         score_threshold: float | None = 0.0,
@@ -573,8 +583,6 @@ class RetrievalService:
         weights: dict | None = None,
         document_ids_filter: list[str] | None = None,
         attachment_id: str | None = None,
-        all_documents: list[Document] = [],
-        exceptions: list[str] = [],
     ):
         if not query and not attachment_id:
             return
@@ -696,3 +704,37 @@ class RetrievalService:
                 }
                 return {"attachment_info": attachment_info, "segment_id": attachment_binding.segment_id}
         return None
+
+    @classmethod
+    def get_segment_attachment_infos(cls, attachment_ids: list[str], session: Session) -> list[dict[str, Any]]:
+        attachment_infos = []
+        upload_files = session.query(UploadFile).where(UploadFile.id.in_(attachment_ids)).all()
+        if upload_files:
+            upload_file_ids = [upload_file.id for upload_file in upload_files]
+            attachment_bindings = (
+                session.query(SegmentAttachmentBinding)
+                .where(SegmentAttachmentBinding.attachment_id.in_(upload_file_ids))
+                .all()
+            )
+            attachment_binding_map = {binding.attachment_id: binding for binding in attachment_bindings}
+
+            if attachment_bindings:
+                for upload_file in upload_files:
+                    attachment_binding = attachment_binding_map.get(upload_file.id)
+                    attachment_info = {
+                        "id": upload_file.id,
+                        "name": upload_file.name,
+                        "extension": "." + upload_file.extension,
+                        "mime_type": upload_file.mime_type,
+                        "source_url": sign_upload_file(upload_file.id, upload_file.extension),
+                        "size": upload_file.size,
+                    }
+                    if attachment_binding:
+                        attachment_infos.append(
+                            {
+                                "attachment_id": attachment_binding.attachment_id,
+                                "attachment_info": attachment_info,
+                                "segment_id": attachment_binding.segment_id,
+                            }
+                        )
+        return attachment_infos

api/core/rag/datasource/vdb/oracle/oraclevector.py

@@ -289,7 +289,8 @@ class OracleVector(BaseVector):
                 words = pseg.cut(query)
                 current_entity = ""
                 for word, pos in words:
-                    if pos in {"nr", "Ng", "eng", "nz", "n", "ORG", "v"}:  # nr: 人名，ns: 地名，nt: 机构名
+                    # `nr`: Person, `ns`: Location, `nt`: Organization
+                    if pos in {"nr", "Ng", "eng", "nz", "n", "ORG", "v"}:
                         current_entity += word
                     else:
                         if current_entity:

api/core/rag/datasource/vdb/pyvastbase/vastbase_vector.py

@@ -213,7 +213,7 @@ class VastbaseVector(BaseVector):
 
             with self._get_cursor() as cur:
                 cur.execute(SQL_CREATE_TABLE.format(table_name=self.table_name, dimension=dimension))
-                # Vastbase 支持的向量维度取值范围为 [1,16000]
+                # Vastbase supports vector dimensions in the range [1, 16,000]
                 if dimension <= 16000:
                     cur.execute(SQL_CREATE_INDEX.format(table_name=self.table_name))
             redis_client.set(collection_exist_cache_key, 1, ex=3600)

api/core/rag/extractor/firecrawl/firecrawl_app.py

@@ -25,7 +25,7 @@ class FirecrawlApp:
         }
         if params:
             json_data.update(params)
-        response = self._post_request(f"{self.base_url}/v2/scrape", json_data, headers)
+        response = self._post_request(self._build_url("v2/scrape"), json_data, headers)
         if response.status_code == 200:
             response_data = response.json()
             data = response_data["data"]
@@ -42,7 +42,7 @@ class FirecrawlApp:
         json_data = {"url": url}
         if params:
             json_data.update(params)
-        response = self._post_request(f"{self.base_url}/v2/crawl", json_data, headers)
+        response = self._post_request(self._build_url("v2/crawl"), json_data, headers)
         if response.status_code == 200:
             # There's also another two fields in the response: "success" (bool) and "url" (str)
             job_id = response.json().get("id")
@@ -58,7 +58,7 @@ class FirecrawlApp:
         if params:
             # Pass through provided params, including optional "sitemap": "only" | "include" | "skip"
             json_data.update(params)
-        response = self._post_request(f"{self.base_url}/v2/map", json_data, headers)
+        response = self._post_request(self._build_url("v2/map"), json_data, headers)
         if response.status_code == 200:
             return cast(dict[str, Any], response.json())
         elif response.status_code in {402, 409, 500, 429, 408}:
@@ -69,7 +69,7 @@ class FirecrawlApp:
 
     def check_crawl_status(self, job_id) -> dict[str, Any]:
         headers = self._prepare_headers()
-        response = self._get_request(f"{self.base_url}/v2/crawl/{job_id}", headers)
+        response = self._get_request(self._build_url(f"v2/crawl/{job_id}"), headers)
         if response.status_code == 200:
             crawl_status_response = response.json()
             if crawl_status_response.get("status") == "completed":
@@ -120,6 +120,10 @@ class FirecrawlApp:
     def _prepare_headers(self) -> dict[str, Any]:
         return {"Content-Type": "application/json", "Authorization": f"Bearer {self.api_key}"}
 
+    def _build_url(self, path: str) -> str:
+        # ensure exactly one slash between base and path, regardless of user-provided base_url
+        return f"{self.base_url.rstrip('/')}/{path.lstrip('/')}"
+
     def _post_request(self, url, data, headers, retries=3, backoff_factor=0.5) -> httpx.Response:
         for attempt in range(retries):
             response = httpx.post(url, headers=headers, json=data)
@@ -139,7 +143,11 @@ class FirecrawlApp:
         return response
 
     def _handle_error(self, response, action):
-        error_message = response.json().get("error", "Unknown error occurred")
+        try:
+            payload = response.json()
+            error_message = payload.get("error") or payload.get("message") or response.text or "Unknown error occurred"
+        except json.JSONDecodeError:
+            error_message = response.text or "Unknown error occurred"
         raise Exception(f"Failed to {action}. Status code: {response.status_code}. Error: {error_message}")  # type: ignore[return]
 
     def search(self, query: str, params: dict[str, Any] | None = None) -> dict[str, Any]:
@@ -160,7 +168,7 @@ class FirecrawlApp:
         }
         if params:
             json_data.update(params)
-        response = self._post_request(f"{self.base_url}/v2/search", json_data, headers)
+        response = self._post_request(self._build_url("v2/search"), json_data, headers)
         if response.status_code == 200:
             response_data = response.json()
             if not response_data.get("success"):

api/core/rag/extractor/notion_extractor.py

@@ -48,13 +48,21 @@ class NotionExtractor(BaseExtractor):
         if notion_access_token:
             self._notion_access_token = notion_access_token
         else:
-            self._notion_access_token = self._get_access_token(tenant_id, self._credential_id)
-            if not self._notion_access_token:
+            try:
+                self._notion_access_token = self._get_access_token(tenant_id, self._credential_id)
+            except Exception as e:
+                logger.warning(
+                    (
+                        "Failed to get Notion access token from datasource credentials: %s, "
+                        "falling back to environment variable NOTION_INTEGRATION_TOKEN"
+                    ),
+                    e,
+                )
                 integration_token = dify_config.NOTION_INTEGRATION_TOKEN
                 if integration_token is None:
                     raise ValueError(
                         "Must specify `integration_token` or set environment variable `NOTION_INTEGRATION_TOKEN`."
-                    )
+                    ) from e
 
                 self._notion_access_token = integration_token
 

api/core/rag/extractor/word_extractor.py

@@ -83,6 +83,7 @@ class WordExtractor(BaseExtractor):
     def _extract_images_from_docx(self, doc):
         image_count = 0
         image_map = {}
+        base_url = dify_config.INTERNAL_FILES_URL or dify_config.FILES_URL
 
         for r_id, rel in doc.part.rels.items():
             if "image" in rel.target_ref:
@@ -121,8 +122,7 @@ class WordExtractor(BaseExtractor):
                             used_at=naive_utc_now(),
                         )
                         db.session.add(upload_file)
-                        # Use r_id as key for external images since target_part is undefined
-                        image_map[r_id] = f"![image]({dify_config.FILES_URL}/files/{upload_file.id}/file-preview)"
+                        image_map[r_id] = f"![image]({base_url}/files/{upload_file.id}/file-preview)"
                 else:
                     image_ext = rel.target_ref.split(".")[-1]
                     if image_ext is None:
@@ -150,10 +150,7 @@ class WordExtractor(BaseExtractor):
                         used_at=naive_utc_now(),
                     )
                     db.session.add(upload_file)
-                    # Use target_part as key for internal images
-                    image_map[rel.target_part] = (
-                        f"![image]({dify_config.FILES_URL}/files/{upload_file.id}/file-preview)"
-                    )
+                    image_map[rel.target_part] = f"![image]({base_url}/files/{upload_file.id}/file-preview)"
         db.session.commit()
         return image_map
 

api/core/rag/index_processor/index_processor_base.py

@@ -231,7 +231,7 @@ class BaseIndexProcessor(ABC):
 
             if not filename:
                 parsed_url = urlparse(image_url)
-                # unquote 处理 URL 中的中文
+                # Decode percent-encoded characters in the URL path.
                 path = unquote(parsed_url.path)
                 filename = os.path.basename(path)
 

api/core/rag/retrieval/dataset_retrieval.py

@@ -151,20 +151,14 @@ class DatasetRetrieval:
             if ModelFeature.TOOL_CALL in features or ModelFeature.MULTI_TOOL_CALL in features:
                 planning_strategy = PlanningStrategy.ROUTER
         available_datasets = []
-        for dataset_id in dataset_ids:
-            # get dataset from dataset id
-            dataset_stmt = select(Dataset).where(Dataset.tenant_id == tenant_id, Dataset.id == dataset_id)
-            dataset = db.session.scalar(dataset_stmt)
 
-            # pass if dataset is not available
-            if not dataset:
+        dataset_stmt = select(Dataset).where(Dataset.tenant_id == tenant_id, Dataset.id.in_(dataset_ids))
+        datasets: list[Dataset] = db.session.execute(dataset_stmt).scalars().all()  # type: ignore
+        for dataset in datasets:
+            if dataset.available_document_count == 0 and dataset.provider != "external":
                 continue
-
-            # pass if dataset is not available
-            if dataset and dataset.available_document_count == 0 and dataset.provider != "external":
-                continue
-
             available_datasets.append(dataset)
+
         if inputs:
             inputs = {key: str(value) for key, value in inputs.items()}
         else:
@@ -282,26 +276,35 @@ class DatasetRetrieval:
                                 )
                                 context_files.append(attachment_info)
                 if show_retrieve_source:
+                    dataset_ids = [record.segment.dataset_id for record in records]
+                    document_ids = [record.segment.document_id for record in records]
+                    dataset_document_stmt = select(DatasetDocument).where(
+                        DatasetDocument.id.in_(document_ids),
+                        DatasetDocument.enabled == True,
+                        DatasetDocument.archived == False,
+                    )
+                    documents = db.session.execute(dataset_document_stmt).scalars().all()  # type: ignore
+                    dataset_stmt = select(Dataset).where(
+                        Dataset.id.in_(dataset_ids),
+                    )
+                    datasets = db.session.execute(dataset_stmt).scalars().all()  # type: ignore
+                    dataset_map = {i.id: i for i in datasets}
+                    document_map = {i.id: i for i in documents}
                     for record in records:
                         segment = record.segment
-                        dataset = db.session.query(Dataset).filter_by(id=segment.dataset_id).first()
-                        dataset_document_stmt = select(DatasetDocument).where(
-                            DatasetDocument.id == segment.document_id,
-                            DatasetDocument.enabled == True,
-                            DatasetDocument.archived == False,
-                        )
-                        document = db.session.scalar(dataset_document_stmt)
-                        if dataset and document:
+                        dataset_item = dataset_map.get(segment.dataset_id)
+                        document_item = document_map.get(segment.document_id)
+                        if dataset_item and document_item:
                             source = RetrievalSourceMetadata(
-                                dataset_id=dataset.id,
-                                dataset_name=dataset.name,
-                                document_id=document.id,
-                                document_name=document.name,
-                                data_source_type=document.data_source_type,
+                                dataset_id=dataset_item.id,
+                                dataset_name=dataset_item.name,
+                                document_id=document_item.id,
+                                document_name=document_item.name,
+                                data_source_type=document_item.data_source_type,
                                 segment_id=segment.id,
                                 retriever_from=invoke_from.to_source(),
                                 score=record.score or 0.0,
-                                doc_metadata=document.doc_metadata,
+                                doc_metadata=document_item.doc_metadata,
                             )
 
                             if invoke_from.to_source() == "dev":

api/core/tools/entities/tool_entities.py

@@ -153,11 +153,11 @@ class ToolInvokeMessage(BaseModel):
         @classmethod
         def transform_variable_value(cls, values):
             """
-            Only basic types and lists are allowed.
+            Only basic types, lists, and None are allowed.
             """
             value = values.get("variable_value")
-            if not isinstance(value, dict | list | str | int | float | bool):
-                raise ValueError("Only basic types and lists are allowed.")
+            if value is not None and not isinstance(value, dict | list | str | int | float | bool):
+                raise ValueError("Only basic types, lists, and None are allowed.")
 
             # if stream is true, the value must be a string
             if values.get("stream"):

api/core/trigger/utils/encryption.py

@@ -67,12 +67,16 @@ def create_trigger_provider_encrypter_for_subscription(
 
 
 def delete_cache_for_subscription(tenant_id: str, provider_id: str, subscription_id: str):
-    cache = TriggerProviderCredentialsCache(
+    TriggerProviderCredentialsCache(
         tenant_id=tenant_id,
         provider_id=provider_id,
         credential_id=subscription_id,
-    )
-    cache.delete()
+    ).delete()
+    TriggerProviderPropertiesCache(
+        tenant_id=tenant_id,
+        provider_id=provider_id,
+        subscription_id=subscription_id,
+    ).delete()
 
 
 def create_trigger_provider_encrypter_for_properties(

api/core/workflow/enums.py

@@ -247,6 +247,7 @@ class WorkflowNodeExecutionMetadataKey(StrEnum):
     ERROR_STRATEGY = "error_strategy"  # node in continue on error mode return the field
     LOOP_VARIABLE_MAP = "loop_variable_map"  # single loop variable output
     DATASOURCE_INFO = "datasource_info"
+    COMPLETED_REASON = "completed_reason"  # completed reason for loop node
 
 
 class WorkflowNodeExecutionStatus(StrEnum):

api/core/workflow/nodes/http_request/executor.py

@@ -86,6 +86,11 @@ class Executor:
             node_data.authorization.config.api_key = variable_pool.convert_template(
                 node_data.authorization.config.api_key
             ).text
+            # Validate that API key is not empty after template conversion
+            if not node_data.authorization.config.api_key or not node_data.authorization.config.api_key.strip():
+                raise AuthorizationConfigError(
+                    "API key is required for authorization but was empty. Please provide a valid API key."
+                )
 
         self.url = node_data.url
         self.method = node_data.method

api/core/workflow/nodes/loop/entities.py

@@ -1,3 +1,4 @@
+from enum import StrEnum
 from typing import Annotated, Any, Literal
 
 from pydantic import AfterValidator, BaseModel, Field, field_validator
@@ -96,3 +97,8 @@ class LoopState(BaseLoopState):
         Get current output.
         """
         return self.current_output
+
+
+class LoopCompletedReason(StrEnum):
+    LOOP_BREAK = "loop_break"
+    LOOP_COMPLETED = "loop_completed"

api/core/workflow/nodes/loop/loop_node.py

@@ -29,7 +29,7 @@ from core.workflow.node_events import (
 )
 from core.workflow.nodes.base import LLMUsageTrackingMixin
 from core.workflow.nodes.base.node import Node
-from core.workflow.nodes.loop.entities import LoopNodeData, LoopVariableData
+from core.workflow.nodes.loop.entities import LoopCompletedReason, LoopNodeData, LoopVariableData
 from core.workflow.utils.condition.processor import ConditionProcessor
 from factories.variable_factory import TypeMismatchError, build_segment_with_type, segment_to_variable
 from libs.datetime_utils import naive_utc_now
@@ -96,6 +96,7 @@ class LoopNode(LLMUsageTrackingMixin, Node[LoopNodeData]):
         loop_duration_map: dict[str, float] = {}
         single_loop_variable_map: dict[str, dict[str, Any]] = {}  # single loop variable output
         loop_usage = LLMUsage.empty_usage()
+        loop_node_ids = self._extract_loop_node_ids_from_config(self.graph_config, self._node_id)
 
         # Start Loop event
         yield LoopStartedEvent(
@@ -118,6 +119,8 @@ class LoopNode(LLMUsageTrackingMixin, Node[LoopNodeData]):
                 loop_count = 0
 
             for i in range(loop_count):
+                # Clear stale variables from previous loop iterations to avoid streaming old values
+                self._clear_loop_subgraph_variables(loop_node_ids)
                 graph_engine = self._create_graph_engine(start_at=start_at, root_node_id=root_node_id)
 
                 loop_start_time = naive_utc_now()
@@ -177,7 +180,11 @@ class LoopNode(LLMUsageTrackingMixin, Node[LoopNodeData]):
                     WorkflowNodeExecutionMetadataKey.TOTAL_TOKENS: loop_usage.total_tokens,
                     WorkflowNodeExecutionMetadataKey.TOTAL_PRICE: loop_usage.total_price,
                     WorkflowNodeExecutionMetadataKey.CURRENCY: loop_usage.currency,
-                    "completed_reason": "loop_break" if reach_break_condition else "loop_completed",
+                    WorkflowNodeExecutionMetadataKey.COMPLETED_REASON: (
+                        LoopCompletedReason.LOOP_BREAK
+                        if reach_break_condition
+                        else LoopCompletedReason.LOOP_COMPLETED.value
+                    ),
                     WorkflowNodeExecutionMetadataKey.LOOP_DURATION_MAP: loop_duration_map,
                     WorkflowNodeExecutionMetadataKey.LOOP_VARIABLE_MAP: single_loop_variable_map,
                 },
@@ -274,6 +281,17 @@ class LoopNode(LLMUsageTrackingMixin, Node[LoopNodeData]):
         if WorkflowNodeExecutionMetadataKey.LOOP_ID not in current_metadata:
             event.node_run_result.metadata = {**current_metadata, **loop_metadata}
 
+    def _clear_loop_subgraph_variables(self, loop_node_ids: set[str]) -> None:
+        """
+        Remove variables produced by loop sub-graph nodes from previous iterations.
+
+        Keeping stale variables causes a freshly created response coordinator in the
+        next iteration to fall back to outdated values when no stream chunks exist.
+        """
+        variable_pool = self.graph_runtime_state.variable_pool
+        for node_id in loop_node_ids:
+            variable_pool.remove([node_id])
+
     @classmethod
     def _extract_variable_selector_to_variable_mapping(
         cls,

api/core/workflow/nodes/parameter_extractor/parameter_extractor_node.py

@@ -281,7 +281,7 @@ class ParameterExtractorNode(Node[ParameterExtractorNodeData]):
 
         # handle invoke result
 
-        text = invoke_result.message.content or ""
+        text = invoke_result.message.get_text_content()
         if not isinstance(text, str):
             raise InvalidTextContentTypeError(f"Invalid text content type: {type(text)}. Expected str.")
 

api/events/event_handlers/__init__.py

@@ -6,6 +6,7 @@ from .create_site_record_when_app_created import handle as handle_create_site_re
 from .delete_tool_parameters_cache_when_sync_draft_workflow import (
     handle as handle_delete_tool_parameters_cache_when_sync_draft_workflow,
 )
+from .queue_credential_sync_when_tenant_created import handle as handle_queue_credential_sync_when_tenant_created
 from .sync_plugin_trigger_when_app_created import handle as handle_sync_plugin_trigger_when_app_created
 from .sync_webhook_when_app_created import handle as handle_sync_webhook_when_app_created
 from .sync_workflow_schedule_when_app_published import handle as handle_sync_workflow_schedule_when_app_published
@@ -30,6 +31,7 @@ __all__ = [
     "handle_create_installed_app_when_app_created",
     "handle_create_site_record_when_app_created",
     "handle_delete_tool_parameters_cache_when_sync_draft_workflow",
+    "handle_queue_credential_sync_when_tenant_created",
     "handle_sync_plugin_trigger_when_app_created",
     "handle_sync_webhook_when_app_created",
     "handle_sync_workflow_schedule_when_app_published",

api/events/event_handlers/queue_credential_sync_when_tenant_created.py

@@ -0,0 +1,19 @@
+from configs import dify_config
+from events.tenant_event import tenant_was_created
+from services.enterprise.workspace_sync import WorkspaceSyncService
+
+
+@tenant_was_created.connect
+def handle(sender, **kwargs):
+    """Queue credential sync when a tenant/workspace is created."""
+    # Only queue sync tasks if plugin manager (enterprise feature) is enabled
+    if not dify_config.ENTERPRISE_ENABLED:
+        return
+
+    tenant = sender
+
+    # Determine source from kwargs if available, otherwise use generic
+    source = kwargs.get("source", "tenant_created")
+
+    # Queue credential sync task to Redis for enterprise backend to process
+    WorkspaceSyncService.queue_credential_sync(tenant.id, source=source)

api/extensions/storage/aliyun_oss_storage.py

@@ -26,6 +26,7 @@ class AliyunOssStorage(BaseStorage):
             self.bucket_name,
             connect_timeout=30,
             region=region,
+            cloudbox_id=dify_config.ALIYUN_CLOUDBOX_ID,
         )
 
     def save(self, filename, data):

api/extensions/storage/huawei_obs_storage.py

@@ -17,6 +17,7 @@ class HuaweiObsStorage(BaseStorage):
             access_key_id=dify_config.HUAWEI_OBS_ACCESS_KEY,
             secret_access_key=dify_config.HUAWEI_OBS_SECRET_KEY,
             server=dify_config.HUAWEI_OBS_SERVER,
+            path_style=dify_config.HUAWEI_OBS_PATH_STYLE,
         )
 
     def save(self, filename, data):

api/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "dify-api"
-version = "1.11.1"
+version = "1.11.2"
 requires-python = ">=3.11,<3.13"
 
 dependencies = [
@@ -69,7 +69,7 @@ dependencies = [
     "pydantic-extra-types~=2.10.3",
     "pydantic-settings~=2.11.0",
     "pyjwt~=2.10.1",
-    "pypdfium2==4.30.0",
+    "pypdfium2==5.2.0",
     "python-docx~=1.1.0",
     "python-dotenv==1.0.1",
     "pyyaml~=6.0.1",

api/services/app_dsl_service.py

@@ -155,6 +155,7 @@ class AppDslService:
                     parsed_url.scheme == "https"
                     and parsed_url.netloc == "github.com"
                     and parsed_url.path.endswith((".yml", ".yaml"))
+                    and "/blob/" in parsed_url.path
                 ):
                     yaml_url = yaml_url.replace("https://github.com", "https://raw.githubusercontent.com")
                     yaml_url = yaml_url.replace("/blob/", "/")

api/services/auth/firecrawl/firecrawl.py

@@ -26,7 +26,7 @@ class FirecrawlAuth(ApiKeyAuthBase):
             "limit": 1,
             "scrapeOptions": {"onlyMainContent": True},
         }
-        response = self._post_request(f"{self.base_url}/v1/crawl", options, headers)
+        response = self._post_request(self._build_url("v1/crawl"), options, headers)
         if response.status_code == 200:
             return True
         else:
@@ -35,15 +35,17 @@ class FirecrawlAuth(ApiKeyAuthBase):
     def _prepare_headers(self):
         return {"Content-Type": "application/json", "Authorization": f"Bearer {self.api_key}"}
 
+    def _build_url(self, path: str) -> str:
+        # ensure exactly one slash between base and path, regardless of user-provided base_url
+        return f"{self.base_url.rstrip('/')}/{path.lstrip('/')}"
+
     def _post_request(self, url, data, headers):
         return httpx.post(url, headers=headers, json=data)
 
     def _handle_error(self, response):
-        if response.status_code in {402, 409, 500}:
-            error_message = response.json().get("error", "Unknown error occurred")
-            raise Exception(f"Failed to authorize. Status code: {response.status_code}. Error: {error_message}")
-        else:
-            if response.text:
-                error_message = json.loads(response.text).get("error", "Unknown error occurred")
-                raise Exception(f"Failed to authorize. Status code: {response.status_code}. Error: {error_message}")
-            raise Exception(f"Unexpected error occurred while trying to authorize. Status code: {response.status_code}")
+        try:
+            payload = response.json()
+        except json.JSONDecodeError:
+            payload = {}
+        error_message = payload.get("error") or payload.get("message") or (response.text or "Unknown error occurred")
+        raise Exception(f"Failed to authorize. Status code: {response.status_code}. Error: {error_message}")

api/services/conversation_service.py

@@ -6,7 +6,9 @@ from typing import Any, Union
 from sqlalchemy import asc, desc, func, or_, select
 from sqlalchemy.orm import Session
 
+from configs import dify_config
 from core.app.entities.app_invoke_entities import InvokeFrom
+from core.db.session_factory import session_factory
 from core.llm_generator.llm_generator import LLMGenerator
 from core.variables.types import SegmentType
 from core.workflow.nodes.variable_assigner.common.impl import conversation_variable_updater_factory
@@ -202,6 +204,7 @@ class ConversationService:
         user: Union[Account, EndUser] | None,
         limit: int,
         last_id: str | None,
+        variable_name: str | None = None,
     ) -> InfiniteScrollPagination:
         conversation = cls.get_conversation(app_model, conversation_id, user)
 
@@ -212,7 +215,25 @@ class ConversationService:
             .order_by(ConversationVariable.created_at)
         )
 
-        with Session(db.engine) as session:
+        # Apply variable_name filter if provided
+        if variable_name:
+            # Filter using JSON extraction to match variable names case-insensitively
+            escaped_variable_name = variable_name.replace("\\", "\\\\").replace("%", "\\%").replace("_", "\\_")
+            # Filter using JSON extraction to match variable names case-insensitively
+            if dify_config.DB_TYPE in ["mysql", "oceanbase", "seekdb"]:
+                stmt = stmt.where(
+                    func.json_extract(ConversationVariable.data, "$.name").ilike(
+                        f"%{escaped_variable_name}%", escape="\\"
+                    )
+                )
+            elif dify_config.DB_TYPE == "postgresql":
+                stmt = stmt.where(
+                    func.json_extract_path_text(ConversationVariable.data, "name").ilike(
+                        f"%{escaped_variable_name}%", escape="\\"
+                    )
+                )
+
+        with session_factory.create_session() as session:
             if last_id:
                 last_variable = session.scalar(stmt.where(ConversationVariable.id == last_id))
                 if not last_variable:
@@ -279,7 +300,7 @@ class ConversationService:
             .where(ConversationVariable.id == variable_id)
         )
 
-        with Session(db.engine) as session:
+        with session_factory.create_session() as session:
             existing_variable = session.scalar(stmt)
             if not existing_variable:
                 raise ConversationVariableNotExistsError()

api/services/dataset_service.py

@@ -3458,7 +3458,7 @@ class SegmentService:
         if keyword:
             query = query.where(DocumentSegment.content.ilike(f"%{keyword}%"))
 
-        query = query.order_by(DocumentSegment.position.asc())
+        query = query.order_by(DocumentSegment.position.asc(), DocumentSegment.id.asc())
         paginated_segments = db.paginate(select=query, page=page, per_page=limit, max_per_page=100, error_out=False)
 
         return paginated_segments.items, paginated_segments.total

api/services/enterprise/enterprise_service.py

@@ -110,5 +110,5 @@ class EnterpriseService:
             if not app_id:
                 raise ValueError("app_id must be provided.")
 
-            body = {"appId": app_id}
-            EnterpriseRequest.send_request("DELETE", "/webapp/clean", json=body)
+            params = {"appId": app_id}
+            EnterpriseRequest.send_request("DELETE", "/webapp/clean", params=params)

api/services/enterprise/workspace_sync.py

@@ -0,0 +1,58 @@
+import json
+import logging
+import uuid
+from datetime import UTC, datetime
+
+from redis import RedisError
+
+from extensions.ext_redis import redis_client
+
+logger = logging.getLogger(__name__)
+
+WORKSPACE_SYNC_QUEUE = "enterprise:workspace:sync:queue"
+WORKSPACE_SYNC_PROCESSING = "enterprise:workspace:sync:processing"
+
+
+class WorkspaceSyncService:
+    """Service to publish workspace sync tasks to Redis queue for enterprise backend consumption"""
+
+    @staticmethod
+    def queue_credential_sync(workspace_id: str, *, source: str) -> bool:
+        """
+        Queue a credential sync task for a newly created workspace.
+
+        This publishes a task to Redis that will be consumed by the enterprise backend
+        worker to sync credentials with the plugin-manager.
+
+        Args:
+            workspace_id: The workspace/tenant ID to sync credentials for
+            source: Source of the sync request (for debugging/tracking)
+
+        Returns:
+            bool: True if task was queued successfully, False otherwise
+        """
+        try:
+            task = {
+                "task_id": str(uuid.uuid4()),
+                "workspace_id": workspace_id,
+                "retry_count": 0,
+                "created_at": datetime.now(UTC).isoformat(),
+                "source": source,
+            }
+
+            # Push to Redis list (queue) - LPUSH adds to the head, worker consumes from tail with RPOP
+            redis_client.lpush(WORKSPACE_SYNC_QUEUE, json.dumps(task))
+
+            logger.info(
+                "Queued credential sync task for workspace %s, task_id: %s, source: %s",
+                workspace_id,
+                task["task_id"],
+                source,
+            )
+            return True
+
+        except (RedisError, TypeError) as e:
+            logger.error("Failed to queue credential sync for workspace %s: %s", workspace_id, str(e), exc_info=True)
+            # Don't raise - we don't want to fail workspace creation if queueing fails
+            # The scheduled task will catch it later
+            return False

api/services/plugin/plugin_parameter_service.py

@@ -105,3 +105,49 @@ class PluginParameterService:
             )
             .options
         )
+
+    @staticmethod
+    def get_dynamic_select_options_with_credentials(
+        tenant_id: str,
+        user_id: str,
+        plugin_id: str,
+        provider: str,
+        action: str,
+        parameter: str,
+        credential_id: str,
+        credentials: Mapping[str, Any],
+    ) -> Sequence[PluginParameterOption]:
+        """
+        Get dynamic select options using provided credentials directly.
+        Used for edit mode when credentials have been modified but not yet saved.
+
+        Security: credential_id is validated against tenant_id to ensure
+        users can only access their own credentials.
+        """
+        from constants import HIDDEN_VALUE
+
+        # Get original subscription to replace hidden values (with tenant_id check for security)
+        original_subscription = TriggerProviderService.get_subscription_by_id(tenant_id, credential_id)
+        if not original_subscription:
+            raise ValueError(f"Subscription {credential_id} not found")
+
+        # Replace [__HIDDEN__] with original values
+        resolved_credentials: dict[str, Any] = {
+            key: (original_subscription.credentials.get(key) if value == HIDDEN_VALUE else value)
+            for key, value in credentials.items()
+        }
+
+        return (
+            DynamicSelectClient()
+            .fetch_dynamic_select_options(
+                tenant_id,
+                user_id,
+                plugin_id,
+                provider,
+                action,
+                resolved_credentials,
+                CredentialType.API_KEY.value,
+                parameter,
+            )
+            .options
+        )

api/services/tools/builtin_tools_manage_service.py

@@ -286,12 +286,12 @@ class BuiltinToolManageService:
 
                     session.add(db_provider)
                     session.commit()
-
-                    # Invalidate tool providers cache
-                    ToolProviderListCache.invalidate_cache(tenant_id)
             except Exception as e:
                 session.rollback()
                 raise ValueError(str(e))
+
+        # Invalidate tool providers cache
+        ToolProviderListCache.invalidate_cache(tenant_id, "builtin")
         return {"result": "success"}
 
     @staticmethod

api/services/tools/mcp_tools_manage_service.py

@@ -15,7 +15,6 @@ from sqlalchemy.orm import Session
 from core.entities.mcp_provider import MCPAuthentication, MCPConfiguration, MCPProviderEntity
 from core.helper import encrypter
 from core.helper.provider_cache import NoOpProviderCredentialCache
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.mcp.auth.auth_flow import auth
 from core.mcp.auth_client import MCPClientWithAuthRetry
 from core.mcp.error import MCPAuthError, MCPError
@@ -65,6 +64,15 @@ class ServerUrlValidationResult(BaseModel):
         return self.needs_validation and self.validation_passed and self.reconnect_result is not None
 
 
+class ProviderUrlValidationData(BaseModel):
+    """Data required for URL validation, extracted from database to perform network operations outside of session"""
+
+    current_server_url_hash: str
+    headers: dict[str, str]
+    timeout: float | None
+    sse_read_timeout: float | None
+
+
 class MCPToolManageService:
     """Service class for managing MCP tools and providers."""
 
@@ -166,9 +174,6 @@ class MCPToolManageService:
         self._session.add(mcp_tool)
         self._session.flush()
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         mcp_providers = ToolTransformService.mcp_provider_to_user_provider(mcp_tool, for_list=True)
         return mcp_providers
 
@@ -192,7 +197,7 @@ class MCPToolManageService:
         Update an MCP provider.
 
         Args:
-            validation_result: Pre-validation result from validate_server_url_change.
+            validation_result: Pre-validation result from validate_server_url_standalone.
                               If provided and contains reconnect_result, it will be used
                               instead of performing network operations.
         """
@@ -251,8 +256,6 @@ class MCPToolManageService:
             # Flush changes to database
             self._session.flush()
 
-            # Invalidate tool providers cache
-            ToolProviderListCache.invalidate_cache(tenant_id)
         except IntegrityError as e:
             self._handle_integrity_error(e, name, server_url, server_identifier)
 
@@ -261,9 +264,6 @@ class MCPToolManageService:
         mcp_tool = self.get_provider(provider_id=provider_id, tenant_id=tenant_id)
         self._session.delete(mcp_tool)
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
     def list_providers(
         self, *, tenant_id: str, for_list: bool = False, include_sensitive: bool = True
     ) -> list[ToolProviderApiEntity]:
@@ -546,30 +546,39 @@ class MCPToolManageService:
         )
         return self.execute_auth_actions(auth_result)
 
-    def _reconnect_provider(self, *, server_url: str, provider: MCPToolProvider) -> ReconnectResult:
-        """Attempt to reconnect to MCP provider with new server URL."""
+    def get_provider_for_url_validation(self, *, tenant_id: str, provider_id: str) -> ProviderUrlValidationData:
+        """
+        Get provider data required for URL validation.
+        This method performs database read and should be called within a session.
+
+        Returns:
+            ProviderUrlValidationData: Data needed for standalone URL validation
+        """
+        provider = self.get_provider(provider_id=provider_id, tenant_id=tenant_id)
         provider_entity = provider.to_entity()
-        headers = provider_entity.headers
+        return ProviderUrlValidationData(
+            current_server_url_hash=provider.server_url_hash,
+            headers=provider_entity.headers,
+            timeout=provider_entity.timeout,
+            sse_read_timeout=provider_entity.sse_read_timeout,
+        )
 
-        try:
-            tools = self._retrieve_remote_mcp_tools(server_url, headers, provider_entity)
-            return ReconnectResult(
-                authed=True,
-                tools=json.dumps([tool.model_dump() for tool in tools]),
-                encrypted_credentials=EMPTY_CREDENTIALS_JSON,
-            )
-        except MCPAuthError:
-            return ReconnectResult(authed=False, tools=EMPTY_TOOLS_JSON, encrypted_credentials=EMPTY_CREDENTIALS_JSON)
-        except MCPError as e:
-            raise ValueError(f"Failed to re-connect MCP server: {e}") from e
-
-    def validate_server_url_change(
-        self, *, tenant_id: str, provider_id: str, new_server_url: str
+    @staticmethod
+    def validate_server_url_standalone(
+        *,
+        tenant_id: str,
+        new_server_url: str,
+        validation_data: ProviderUrlValidationData,
     ) -> ServerUrlValidationResult:
         """
         Validate server URL change by attempting to connect to the new server.
-        This method should be called BEFORE update_provider to perform network operations
-        outside of the database transaction.
+        This method performs network operations and MUST be called OUTSIDE of any database session
+        to avoid holding locks during network I/O.
+
+        Args:
+            tenant_id: Tenant ID for encryption
+            new_server_url: The new server URL to validate
+            validation_data: Provider data obtained from get_provider_for_url_validation
 
         Returns:
             ServerUrlValidationResult: Validation result with connection status and tools if successful
@@ -579,25 +588,30 @@ class MCPToolManageService:
             return ServerUrlValidationResult(needs_validation=False)
 
         # Validate URL format
-        if not self._is_valid_url(new_server_url):
+        parsed = urlparse(new_server_url)
+        if not all([parsed.scheme, parsed.netloc]) or parsed.scheme not in ["http", "https"]:
             raise ValueError("Server URL is not valid.")
 
         # Always encrypt and hash the URL
         encrypted_server_url = encrypter.encrypt_token(tenant_id, new_server_url)
         new_server_url_hash = hashlib.sha256(new_server_url.encode()).hexdigest()
 
-        # Get current provider
-        provider = self.get_provider(provider_id=provider_id, tenant_id=tenant_id)
-
         # Check if URL is actually different
-        if new_server_url_hash == provider.server_url_hash:
+        if new_server_url_hash == validation_data.current_server_url_hash:
             # URL hasn't changed, but still return the encrypted data
             return ServerUrlValidationResult(
-                needs_validation=False, encrypted_server_url=encrypted_server_url, server_url_hash=new_server_url_hash
+                needs_validation=False,
+                encrypted_server_url=encrypted_server_url,
+                server_url_hash=new_server_url_hash,
             )
 
-        # Perform validation by attempting to connect
-        reconnect_result = self._reconnect_provider(server_url=new_server_url, provider=provider)
+        # Perform network validation - this is the expensive operation that should be outside session
+        reconnect_result = MCPToolManageService._reconnect_with_url(
+            server_url=new_server_url,
+            headers=validation_data.headers,
+            timeout=validation_data.timeout,
+            sse_read_timeout=validation_data.sse_read_timeout,
+        )
         return ServerUrlValidationResult(
             needs_validation=True,
             validation_passed=True,
@@ -606,6 +620,38 @@ class MCPToolManageService:
             server_url_hash=new_server_url_hash,
         )
 
+    @staticmethod
+    def _reconnect_with_url(
+        *,
+        server_url: str,
+        headers: dict[str, str],
+        timeout: float | None,
+        sse_read_timeout: float | None,
+    ) -> ReconnectResult:
+        """
+        Attempt to connect to MCP server with given URL.
+        This is a static method that performs network I/O without database access.
+        """
+        from core.mcp.mcp_client import MCPClient
+
+        try:
+            with MCPClient(
+                server_url=server_url,
+                headers=headers,
+                timeout=timeout,
+                sse_read_timeout=sse_read_timeout,
+            ) as mcp_client:
+                tools = mcp_client.list_tools()
+                return ReconnectResult(
+                    authed=True,
+                    tools=json.dumps([tool.model_dump() for tool in tools]),
+                    encrypted_credentials=EMPTY_CREDENTIALS_JSON,
+                )
+        except MCPAuthError:
+            return ReconnectResult(authed=False, tools=EMPTY_TOOLS_JSON, encrypted_credentials=EMPTY_CREDENTIALS_JSON)
+        except MCPError as e:
+            raise ValueError(f"Failed to re-connect MCP server: {e}") from e
+
     def _build_tool_provider_response(
         self, db_provider: MCPToolProvider, provider_entity: MCPProviderEntity, tools: list
     ) -> ToolProviderApiEntity:

api/services/trigger/trigger_provider_service.py

@@ -94,16 +94,23 @@ class TriggerProviderService:
 
         provider_controller = TriggerManager.get_trigger_provider(tenant_id, provider_id)
         for subscription in subscriptions:
-            encrypter, _ = create_trigger_provider_encrypter_for_subscription(
+            credential_encrypter, _ = create_trigger_provider_encrypter_for_subscription(
                 tenant_id=tenant_id,
                 controller=provider_controller,
                 subscription=subscription,
             )
             subscription.credentials = dict(
-                encrypter.mask_credentials(dict(encrypter.decrypt(subscription.credentials)))
+                credential_encrypter.mask_credentials(dict(credential_encrypter.decrypt(subscription.credentials)))
             )
-            subscription.properties = dict(encrypter.mask_credentials(dict(encrypter.decrypt(subscription.properties))))
-            subscription.parameters = dict(encrypter.mask_credentials(dict(encrypter.decrypt(subscription.parameters))))
+            properties_encrypter, _ = create_trigger_provider_encrypter_for_properties(
+                tenant_id=tenant_id,
+                controller=provider_controller,
+                subscription=subscription,
+            )
+            subscription.properties = dict(
+                properties_encrypter.mask_credentials(dict(properties_encrypter.decrypt(subscription.properties)))
+            )
+            subscription.parameters = dict(subscription.parameters)
             count = workflows_in_use_map.get(subscription.id)
             subscription.workflows_in_use = count if count is not None else 0
 
@@ -209,6 +216,101 @@ class TriggerProviderService:
             logger.exception("Failed to add trigger provider")
             raise ValueError(str(e))
 
+    @classmethod
+    def update_trigger_subscription(
+        cls,
+        tenant_id: str,
+        subscription_id: str,
+        name: str | None = None,
+        properties: Mapping[str, Any] | None = None,
+        parameters: Mapping[str, Any] | None = None,
+        credentials: Mapping[str, Any] | None = None,
+        credential_expires_at: int | None = None,
+        expires_at: int | None = None,
+    ) -> None:
+        """
+        Update an existing trigger subscription.
+
+        :param tenant_id: Tenant ID
+        :param subscription_id: Subscription instance ID
+        :param name: Optional new name for this subscription
+        :param properties: Optional new properties
+        :param parameters: Optional new parameters
+        :param credentials: Optional new credentials
+        :param credential_expires_at: Optional new credential expiration timestamp
+        :param expires_at: Optional new expiration timestamp
+        :return: Success response with updated subscription info
+        """
+        with Session(db.engine, expire_on_commit=False) as session:
+            # Use distributed lock to prevent race conditions on the same subscription
+            lock_key = f"trigger_subscription_update_lock:{tenant_id}_{subscription_id}"
+            with redis_client.lock(lock_key, timeout=20):
+                subscription: TriggerSubscription | None = (
+                    session.query(TriggerSubscription).filter_by(tenant_id=tenant_id, id=subscription_id).first()
+                )
+                if not subscription:
+                    raise ValueError(f"Trigger subscription {subscription_id} not found")
+
+                provider_id = TriggerProviderID(subscription.provider_id)
+                provider_controller = TriggerManager.get_trigger_provider(tenant_id, provider_id)
+
+                # Check for name uniqueness if name is being updated
+                if name is not None and name != subscription.name:
+                    existing = (
+                        session.query(TriggerSubscription)
+                        .filter_by(tenant_id=tenant_id, provider_id=str(provider_id), name=name)
+                        .first()
+                    )
+                    if existing:
+                        raise ValueError(f"Subscription name '{name}' already exists for this provider")
+                    subscription.name = name
+
+                # Update properties if provided
+                if properties is not None:
+                    properties_encrypter, _ = create_provider_encrypter(
+                        tenant_id=tenant_id,
+                        config=provider_controller.get_properties_schema(),
+                        cache=NoOpProviderCredentialCache(),
+                    )
+                    # Handle hidden values - preserve original encrypted values
+                    original_properties = properties_encrypter.decrypt(subscription.properties)
+                    new_properties: dict[str, Any] = {
+                        key: value if value != HIDDEN_VALUE else original_properties.get(key, UNKNOWN_VALUE)
+                        for key, value in properties.items()
+                    }
+                    subscription.properties = dict(properties_encrypter.encrypt(new_properties))
+
+                # Update parameters if provided
+                if parameters is not None:
+                    subscription.parameters = dict(parameters)
+
+                # Update credentials if provided
+                if credentials is not None:
+                    credential_type = CredentialType.of(subscription.credential_type)
+                    credential_encrypter, _ = create_provider_encrypter(
+                        tenant_id=tenant_id,
+                        config=provider_controller.get_credential_schema_config(credential_type),
+                        cache=NoOpProviderCredentialCache(),
+                    )
+                    subscription.credentials = dict(credential_encrypter.encrypt(dict(credentials)))
+
+                # Update credential expiration timestamp if provided
+                if credential_expires_at is not None:
+                    subscription.credential_expires_at = credential_expires_at
+
+                # Update expiration timestamp if provided
+                if expires_at is not None:
+                    subscription.expires_at = expires_at
+
+                session.commit()
+
+                # Clear subscription cache
+                delete_cache_for_subscription(
+                    tenant_id=tenant_id,
+                    provider_id=subscription.provider_id,
+                    subscription_id=subscription.id,
+                )
+
     @classmethod
     def get_subscription_by_id(cls, tenant_id: str, subscription_id: str | None = None) -> TriggerSubscription | None:
         """
@@ -257,17 +359,18 @@ class TriggerProviderService:
             raise ValueError(f"Trigger provider subscription {subscription_id} not found")
 
         credential_type: CredentialType = CredentialType.of(subscription.credential_type)
+        provider_id = TriggerProviderID(subscription.provider_id)
+        provider_controller: PluginTriggerProviderController = TriggerManager.get_trigger_provider(
+            tenant_id=tenant_id, provider_id=provider_id
+        )
+        encrypter, _ = create_trigger_provider_encrypter_for_subscription(
+            tenant_id=tenant_id,
+            controller=provider_controller,
+            subscription=subscription,
+        )
+
         is_auto_created: bool = credential_type in [CredentialType.OAUTH2, CredentialType.API_KEY]
         if is_auto_created:
-            provider_id = TriggerProviderID(subscription.provider_id)
-            provider_controller: PluginTriggerProviderController = TriggerManager.get_trigger_provider(
-                tenant_id=tenant_id, provider_id=provider_id
-            )
-            encrypter, _ = create_trigger_provider_encrypter_for_subscription(
-                tenant_id=tenant_id,
-                controller=provider_controller,
-                subscription=subscription,
-            )
             try:
                 TriggerManager.unsubscribe_trigger(
                     tenant_id=tenant_id,
@@ -280,8 +383,8 @@ class TriggerProviderService:
             except Exception as e:
                 logger.exception("Error unsubscribing trigger", exc_info=e)
 
-        # Clear cache
         session.delete(subscription)
+        # Clear cache
         delete_cache_for_subscription(
             tenant_id=tenant_id,
             provider_id=subscription.provider_id,
@@ -688,3 +791,125 @@ class TriggerProviderService:
             )
             subscription.properties = dict(properties_encrypter.decrypt(subscription.properties))
             return subscription
+
+    @classmethod
+    def verify_subscription_credentials(
+        cls,
+        tenant_id: str,
+        user_id: str,
+        provider_id: TriggerProviderID,
+        subscription_id: str,
+        credentials: Mapping[str, Any],
+    ) -> dict[str, Any]:
+        """
+        Verify credentials for an existing subscription without updating it.
+
+        This is used in edit mode to validate new credentials before rebuild.
+
+        :param tenant_id: Tenant ID
+        :param user_id: User ID
+        :param provider_id: Provider identifier
+        :param subscription_id: Subscription ID
+        :param credentials: New credentials to verify
+        :return: dict with 'verified' boolean
+        """
+        provider_controller = TriggerManager.get_trigger_provider(tenant_id, provider_id)
+        if not provider_controller:
+            raise ValueError(f"Provider {provider_id} not found")
+
+        subscription = cls.get_subscription_by_id(
+            tenant_id=tenant_id,
+            subscription_id=subscription_id,
+        )
+        if not subscription:
+            raise ValueError(f"Subscription {subscription_id} not found")
+
+        credential_type = CredentialType.of(subscription.credential_type)
+
+        # For API Key, validate the new credentials
+        if credential_type == CredentialType.API_KEY:
+            new_credentials: dict[str, Any] = {
+                key: value if value != HIDDEN_VALUE else subscription.credentials.get(key, UNKNOWN_VALUE)
+                for key, value in credentials.items()
+            }
+            try:
+                provider_controller.validate_credentials(user_id, credentials=new_credentials)
+                return {"verified": True}
+            except Exception as e:
+                raise ValueError(f"Invalid credentials: {e}") from e
+
+        return {"verified": True}
+
+    @classmethod
+    def rebuild_trigger_subscription(
+        cls,
+        tenant_id: str,
+        provider_id: TriggerProviderID,
+        subscription_id: str,
+        credentials: Mapping[str, Any],
+        parameters: Mapping[str, Any],
+        name: str | None = None,
+    ) -> None:
+        """
+        Create a subscription builder for rebuilding an existing subscription.
+
+        This method creates a builder pre-filled with data from the rebuild request,
+        keeping the same subscription_id and endpoint_id so the webhook URL remains unchanged.
+
+        :param tenant_id: Tenant ID
+        :param name: Name for the subscription
+        :param subscription_id: Subscription ID
+        :param provider_id: Provider identifier
+        :param credentials: Credentials for the subscription
+        :param parameters: Parameters for the subscription
+        :return: SubscriptionBuilderApiEntity
+        """
+        provider_controller = TriggerManager.get_trigger_provider(tenant_id, provider_id)
+        if not provider_controller:
+            raise ValueError(f"Provider {provider_id} not found")
+
+        subscription = TriggerProviderService.get_subscription_by_id(
+            tenant_id=tenant_id,
+            subscription_id=subscription_id,
+        )
+        if not subscription:
+            raise ValueError(f"Subscription {subscription_id} not found")
+
+        credential_type = CredentialType.of(subscription.credential_type)
+        if credential_type not in [CredentialType.OAUTH2, CredentialType.API_KEY]:
+            raise ValueError("Credential type not supported for rebuild")
+
+        # TODO: Trying to invoke update api of the plugin trigger provider
+
+        # FALLBACK: If the update api is not implemented, delete the previous subscription and create a new one
+
+        # Delete the previous subscription
+        user_id = subscription.user_id
+        TriggerManager.unsubscribe_trigger(
+            tenant_id=tenant_id,
+            user_id=user_id,
+            provider_id=provider_id,
+            subscription=subscription.to_entity(),
+            credentials=subscription.credentials,
+            credential_type=credential_type,
+        )
+
+        # Create a new subscription with the same subscription_id and endpoint_id
+        new_subscription: TriggerSubscriptionEntity = TriggerManager.subscribe_trigger(
+            tenant_id=tenant_id,
+            user_id=user_id,
+            provider_id=provider_id,
+            endpoint=generate_plugin_trigger_endpoint_url(subscription.endpoint_id),
+            parameters=parameters,
+            credentials=credentials,
+            credential_type=credential_type,
+        )
+        TriggerProviderService.update_trigger_subscription(
+            tenant_id=tenant_id,
+            subscription_id=subscription.id,
+            name=name,
+            parameters=parameters,
+            credentials=credentials,
+            properties=new_subscription.properties,
+            expires_at=new_subscription.expires_at,
+        )

api/services/trigger/trigger_subscription_builder_service.py

@@ -453,11 +453,12 @@ class TriggerSubscriptionBuilderService:
         if not subscription_builder:
             return None
 
-        # response to validation endpoint
-        controller: PluginTriggerProviderController = TriggerManager.get_trigger_provider(
-            tenant_id=subscription_builder.tenant_id, provider_id=TriggerProviderID(subscription_builder.provider_id)
-        )
         try:
+            # response to validation endpoint
+            controller: PluginTriggerProviderController = TriggerManager.get_trigger_provider(
+                tenant_id=subscription_builder.tenant_id,
+                provider_id=TriggerProviderID(subscription_builder.provider_id),
+            )
             dispatch_response: TriggerDispatchResponse = controller.dispatch(
                 request=request,
                 subscription=subscription_builder.to_subscription(),

api/tasks/document_indexing_sync_task.py

@@ -2,7 +2,6 @@ import logging
 import time
 
 import click
-import sqlalchemy as sa
 from celery import shared_task
 from sqlalchemy import select
 
@@ -12,7 +11,7 @@ from core.rag.index_processor.index_processor_factory import IndexProcessorFacto
 from extensions.ext_database import db
 from libs.datetime_utils import naive_utc_now
 from models.dataset import Dataset, Document, DocumentSegment
-from models.source import DataSourceOauthBinding
+from services.datasource_provider_service import DatasourceProviderService
 
 logger = logging.getLogger(__name__)
 
@@ -48,27 +47,36 @@ def document_indexing_sync_task(dataset_id: str, document_id: str):
         page_id = data_source_info["notion_page_id"]
         page_type = data_source_info["type"]
         page_edited_time = data_source_info["last_edited_time"]
+        credential_id = data_source_info.get("credential_id")
 
-        data_source_binding = (
-            db.session.query(DataSourceOauthBinding)
-            .where(
-                sa.and_(
-                    DataSourceOauthBinding.tenant_id == document.tenant_id,
-                    DataSourceOauthBinding.provider == "notion",
-                    DataSourceOauthBinding.disabled == False,
-                    DataSourceOauthBinding.source_info["workspace_id"] == f'"{workspace_id}"',
-                )
-            )
-            .first()
+        # Get credentials from datasource provider
+        datasource_provider_service = DatasourceProviderService()
+        credential = datasource_provider_service.get_datasource_credentials(
+            tenant_id=document.tenant_id,
+            credential_id=credential_id,
+            provider="notion_datasource",
+            plugin_id="langgenius/notion_datasource",
         )
-        if not data_source_binding:
-            raise ValueError("Data source binding not found.")
+
+        if not credential:
+            logger.error(
+                "Datasource credential not found for document %s, tenant_id: %s, credential_id: %s",
+                document_id,
+                document.tenant_id,
+                credential_id,
+            )
+            document.indexing_status = "error"
+            document.error = "Datasource credential not found. Please reconnect your Notion workspace."
+            document.stopped_at = naive_utc_now()
+            db.session.commit()
+            db.session.close()
+            return
 
         loader = NotionExtractor(
             notion_workspace_id=workspace_id,
             notion_obj_id=page_id,
             notion_page_type=page_type,
-            notion_access_token=data_source_binding.access_token,
+            notion_access_token=credential.get("integration_secret"),
             tenant_id=document.tenant_id,
         )
 

api/tests/integration_tests/workflow/nodes/test_http.py

@@ -6,6 +6,7 @@ import pytest
 
 from core.app.entities.app_invoke_entities import InvokeFrom
 from core.workflow.entities import GraphInitParams
+from core.workflow.enums import WorkflowNodeExecutionStatus
 from core.workflow.graph import Graph
 from core.workflow.nodes.http_request.node import HttpRequestNode
 from core.workflow.nodes.node_factory import DifyNodeFactory
@@ -169,13 +170,14 @@ def test_custom_authorization_header(setup_http_mock):
 
 
 @pytest.mark.parametrize("setup_http_mock", [["none"]], indirect=True)
-def test_custom_auth_with_empty_api_key_does_not_set_header(setup_http_mock):
-    """Test: In custom authentication mode, when the api_key is empty, no header should be set."""
+def test_custom_auth_with_empty_api_key_raises_error(setup_http_mock):
+    """Test: In custom authentication mode, when the api_key is empty, AuthorizationConfigError should be raised."""
     from core.workflow.nodes.http_request.entities import (
         HttpRequestNodeAuthorization,
         HttpRequestNodeData,
         HttpRequestNodeTimeout,
     )
+    from core.workflow.nodes.http_request.exc import AuthorizationConfigError
     from core.workflow.nodes.http_request.executor import Executor
     from core.workflow.runtime import VariablePool
     from core.workflow.system_variable import SystemVariable
@@ -208,16 +210,13 @@ def test_custom_auth_with_empty_api_key_does_not_set_header(setup_http_mock):
         ssl_verify=True,
     )
 
-    # Create executor
-    executor = Executor(
-        node_data=node_data, timeout=HttpRequestNodeTimeout(connect=10, read=30, write=10), variable_pool=variable_pool
-    )
-
-    # Get assembled headers
-    headers = executor._assembling_headers()
-
-    # When api_key is empty, the custom header should NOT be set
-    assert "X-Custom-Auth" not in headers
+    # Create executor should raise AuthorizationConfigError
+    with pytest.raises(AuthorizationConfigError, match="API key is required"):
+        Executor(
+            node_data=node_data,
+            timeout=HttpRequestNodeTimeout(connect=10, read=30, write=10),
+            variable_pool=variable_pool,
+        )
 
 
 @pytest.mark.parametrize("setup_http_mock", [["none"]], indirect=True)
@@ -305,9 +304,10 @@ def test_basic_authorization_with_custom_header_ignored(setup_http_mock):
 @pytest.mark.parametrize("setup_http_mock", [["none"]], indirect=True)
 def test_custom_authorization_with_empty_api_key(setup_http_mock):
     """
-    Test that custom authorization doesn't set header when api_key is empty.
-    This test verifies the fix for issue #23554.
+    Test that custom authorization raises error when api_key is empty.
+    This test verifies the fix for issue #21830.
     """
+
     node = init_http_node(
         config={
             "id": "1",
@@ -333,11 +333,10 @@ def test_custom_authorization_with_empty_api_key(setup_http_mock):
     )
 
     result = node._run()
-    assert result.process_data is not None
-    data = result.process_data.get("request", "")
-
-    # Custom header should NOT be set when api_key is empty
-    assert "X-Custom-Auth:" not in data
+    # Should fail with AuthorizationConfigError
+    assert result.status == WorkflowNodeExecutionStatus.FAILED
+    assert "API key is required" in result.error
+    assert result.error_type == "AuthorizationConfigError"
 
 
 @pytest.mark.parametrize("setup_http_mock", [["none"]], indirect=True)

api/tests/test_containers_integration_tests/services/tools/test_mcp_tools_manage_service.py

@@ -1308,18 +1308,17 @@ class TestMCPToolManageService:
             type("MockTool", (), {"model_dump": lambda self: {"name": "test_tool_2", "description": "Test tool 2"}})(),
         ]
 
-        with patch("services.tools.mcp_tools_manage_service.MCPClientWithAuthRetry") as mock_mcp_client:
+        with patch("core.mcp.mcp_client.MCPClient") as mock_mcp_client:
             # Setup mock client
             mock_client_instance = mock_mcp_client.return_value.__enter__.return_value
             mock_client_instance.list_tools.return_value = mock_tools
 
             # Act: Execute the method under test
-            from extensions.ext_database import db
-
-            service = MCPToolManageService(db.session())
-            result = service._reconnect_provider(
+            result = MCPToolManageService._reconnect_with_url(
                 server_url="https://example.com/mcp",
-                provider=mcp_provider,
+                headers={"X-Test": "1"},
+                timeout=mcp_provider.timeout,
+                sse_read_timeout=mcp_provider.sse_read_timeout,
             )
 
         # Assert: Verify the expected outcomes
@@ -1337,8 +1336,12 @@ class TestMCPToolManageService:
         assert tools_data[1]["name"] == "test_tool_2"
 
         # Verify mock interactions
-        provider_entity = mcp_provider.to_entity()
-        mock_mcp_client.assert_called_once()
+        mock_mcp_client.assert_called_once_with(
+            server_url="https://example.com/mcp",
+            headers={"X-Test": "1"},
+            timeout=mcp_provider.timeout,
+            sse_read_timeout=mcp_provider.sse_read_timeout,
+        )
 
     def test_re_connect_mcp_provider_auth_error(self, db_session_with_containers, mock_external_service_dependencies):
         """
@@ -1361,19 +1364,18 @@ class TestMCPToolManageService:
         )
 
         # Mock MCPClient to raise authentication error
-        with patch("services.tools.mcp_tools_manage_service.MCPClientWithAuthRetry") as mock_mcp_client:
+        with patch("core.mcp.mcp_client.MCPClient") as mock_mcp_client:
             from core.mcp.error import MCPAuthError
 
             mock_client_instance = mock_mcp_client.return_value.__enter__.return_value
             mock_client_instance.list_tools.side_effect = MCPAuthError("Authentication required")
 
             # Act: Execute the method under test
-            from extensions.ext_database import db
-
-            service = MCPToolManageService(db.session())
-            result = service._reconnect_provider(
+            result = MCPToolManageService._reconnect_with_url(
                 server_url="https://example.com/mcp",
-                provider=mcp_provider,
+                headers={},
+                timeout=mcp_provider.timeout,
+                sse_read_timeout=mcp_provider.sse_read_timeout,
             )
 
         # Assert: Verify the expected outcomes
@@ -1404,18 +1406,17 @@ class TestMCPToolManageService:
         )
 
         # Mock MCPClient to raise connection error
-        with patch("services.tools.mcp_tools_manage_service.MCPClientWithAuthRetry") as mock_mcp_client:
+        with patch("core.mcp.mcp_client.MCPClient") as mock_mcp_client:
             from core.mcp.error import MCPError
 
             mock_client_instance = mock_mcp_client.return_value.__enter__.return_value
             mock_client_instance.list_tools.side_effect = MCPError("Connection failed")
 
             # Act & Assert: Verify proper error handling
-            from extensions.ext_database import db
-
-            service = MCPToolManageService(db.session())
             with pytest.raises(ValueError, match="Failed to re-connect MCP server: Connection failed"):
-                service._reconnect_provider(
+                MCPToolManageService._reconnect_with_url(
                     server_url="https://example.com/mcp",
-                    provider=mcp_provider,
+                    headers={"X-Test": "1"},
+                    timeout=mcp_provider.timeout,
+                    sse_read_timeout=mcp_provider.sse_read_timeout,
                 )

api/tests/unit_tests/controllers/common/test_file_response.py

@@ -0,0 +1,46 @@
+from flask import Response
+
+from controllers.common.file_response import enforce_download_for_html, is_html_content
+
+
+class TestFileResponseHelpers:
+    def test_is_html_content_detects_mime_type(self):
+        mime_type = "text/html; charset=UTF-8"
+
+        result = is_html_content(mime_type, filename="file.txt", extension="txt")
+
+        assert result is True
+
+    def test_is_html_content_detects_extension(self):
+        result = is_html_content("text/plain", filename="report.html", extension=None)
+
+        assert result is True
+
+    def test_enforce_download_for_html_sets_headers(self):
+        response = Response("payload", mimetype="text/html")
+
+        updated = enforce_download_for_html(
+            response,
+            mime_type="text/html",
+            filename="unsafe.html",
+            extension="html",
+        )
+
+        assert updated is True
+        assert "attachment" in response.headers["Content-Disposition"]
+        assert response.headers["Content-Type"] == "application/octet-stream"
+        assert response.headers["X-Content-Type-Options"] == "nosniff"
+
+    def test_enforce_download_for_html_no_change_for_non_html(self):
+        response = Response("payload", mimetype="text/plain")
+
+        updated = enforce_download_for_html(
+            response,
+            mime_type="text/plain",
+            filename="notes.txt",
+            extension="txt",
+        )
+
+        assert updated is False
+        assert "Content-Disposition" not in response.headers
+        assert "X-Content-Type-Options" not in response.headers

api/tests/unit_tests/controllers/console/auth/test_account_activation.py

@@ -163,34 +163,17 @@ class TestActivateApi:
             "account": mock_account,
         }
 
-    @pytest.fixture
-    def mock_token_pair(self):
-        """Create mock token pair object."""
-        token_pair = MagicMock()
-        token_pair.access_token = "access_token"
-        token_pair.refresh_token = "refresh_token"
-        token_pair.csrf_token = "csrf_token"
-        token_pair.model_dump.return_value = {
-            "access_token": "access_token",
-            "refresh_token": "refresh_token",
-            "csrf_token": "csrf_token",
-        }
-        return token_pair
-
     @patch("controllers.console.auth.activate.RegisterService.get_invitation_if_token_valid")
     @patch("controllers.console.auth.activate.RegisterService.revoke_token")
     @patch("controllers.console.auth.activate.db")
-    @patch("controllers.console.auth.activate.AccountService.login")
     def test_successful_account_activation(
         self,
-        mock_login,
         mock_db,
         mock_revoke_token,
         mock_get_invitation,
         app,
         mock_invitation,
         mock_account,
-        mock_token_pair,
     ):
         """
         Test successful account activation.
@@ -198,12 +181,10 @@ class TestActivateApi:
         Verifies that:
         - Account is activated with user preferences
         - Account status is set to ACTIVE
-        - User is logged in after activation
         - Invitation token is revoked
         """
         # Arrange
         mock_get_invitation.return_value = mock_invitation
-        mock_login.return_value = mock_token_pair
 
         # Act
         with app.test_request_context(
@@ -230,7 +211,6 @@ class TestActivateApi:
         assert mock_account.initialized_at is not None
         mock_revoke_token.assert_called_once_with("workspace-123", "[email protected]", "valid_token")
         mock_db.session.commit.assert_called_once()
-        mock_login.assert_called_once()
 
     @patch("controllers.console.auth.activate.RegisterService.get_invitation_if_token_valid")
     def test_activation_with_invalid_token(self, mock_get_invitation, app):
@@ -264,17 +244,14 @@ class TestActivateApi:
     @patch("controllers.console.auth.activate.RegisterService.get_invitation_if_token_valid")
     @patch("controllers.console.auth.activate.RegisterService.revoke_token")
     @patch("controllers.console.auth.activate.db")
-    @patch("controllers.console.auth.activate.AccountService.login")
     def test_activation_sets_interface_theme(
         self,
-        mock_login,
         mock_db,
         mock_revoke_token,
         mock_get_invitation,
         app,
         mock_invitation,
         mock_account,
-        mock_token_pair,
     ):
         """
         Test that activation sets default interface theme.
@@ -284,7 +261,6 @@ class TestActivateApi:
         """
         # Arrange
         mock_get_invitation.return_value = mock_invitation
-        mock_login.return_value = mock_token_pair
 
         # Act
         with app.test_request_context(
@@ -317,17 +293,14 @@ class TestActivateApi:
     @patch("controllers.console.auth.activate.RegisterService.get_invitation_if_token_valid")
     @patch("controllers.console.auth.activate.RegisterService.revoke_token")
     @patch("controllers.console.auth.activate.db")
-    @patch("controllers.console.auth.activate.AccountService.login")
     def test_activation_with_different_locales(
         self,
-        mock_login,
         mock_db,
         mock_revoke_token,
         mock_get_invitation,
         app,
         mock_invitation,
         mock_account,
-        mock_token_pair,
         language,
         timezone,
     ):
@@ -341,7 +314,6 @@ class TestActivateApi:
         """
         # Arrange
         mock_get_invitation.return_value = mock_invitation
-        mock_login.return_value = mock_token_pair
 
         # Act
         with app.test_request_context(
@@ -367,27 +339,23 @@ class TestActivateApi:
     @patch("controllers.console.auth.activate.RegisterService.get_invitation_if_token_valid")
     @patch("controllers.console.auth.activate.RegisterService.revoke_token")
     @patch("controllers.console.auth.activate.db")
-    @patch("controllers.console.auth.activate.AccountService.login")
-    def test_activation_returns_token_data(
+    def test_activation_returns_success_response(
         self,
-        mock_login,
         mock_db,
         mock_revoke_token,
         mock_get_invitation,
         app,
         mock_invitation,
-        mock_token_pair,
     ):
         """
-        Test that activation returns authentication tokens.
+        Test that activation returns a success response without authentication tokens.
 
         Verifies that:
-        - Token pair is returned in response
-        - All token types are included (access, refresh, csrf)
+        - Response contains a success result
+        - No token data is returned
         """
         # Arrange
         mock_get_invitation.return_value = mock_invitation
-        mock_login.return_value = mock_token_pair
 
         # Act
         with app.test_request_context(
@@ -406,24 +374,18 @@ class TestActivateApi:
             response = api.post()
 
         # Assert
-        assert "data" in response
-        assert response["data"]["access_token"] == "access_token"
-        assert response["data"]["refresh_token"] == "refresh_token"
-        assert response["data"]["csrf_token"] == "csrf_token"
+        assert response == {"result": "success"}
 
     @patch("controllers.console.auth.activate.RegisterService.get_invitation_if_token_valid")
     @patch("controllers.console.auth.activate.RegisterService.revoke_token")
     @patch("controllers.console.auth.activate.db")
-    @patch("controllers.console.auth.activate.AccountService.login")
     def test_activation_without_workspace_id(
         self,
-        mock_login,
         mock_db,
         mock_revoke_token,
         mock_get_invitation,
         app,
         mock_invitation,
-        mock_token_pair,
     ):
         """
         Test account activation without workspace_id.
@@ -434,7 +396,6 @@ class TestActivateApi:
         """
         # Arrange
         mock_get_invitation.return_value = mock_invitation
-        mock_login.return_value = mock_token_pair
 
         # Act
         with app.test_request_context(

api/tests/unit_tests/controllers/console/test_extension.py

@@ -0,0 +1,236 @@
+from __future__ import annotations
+
+import builtins
+import uuid
+from datetime import UTC, datetime
+from unittest.mock import MagicMock
+
+import pytest
+from flask import Flask
+from flask.views import MethodView as FlaskMethodView
+
+_NEEDS_METHOD_VIEW_CLEANUP = False
+if not hasattr(builtins, "MethodView"):
+    builtins.MethodView = FlaskMethodView
+    _NEEDS_METHOD_VIEW_CLEANUP = True
+
+from constants import HIDDEN_VALUE
+from controllers.console.extension import (
+    APIBasedExtensionAPI,
+    APIBasedExtensionDetailAPI,
+    CodeBasedExtensionAPI,
+)
+
+if _NEEDS_METHOD_VIEW_CLEANUP:
+    delattr(builtins, "MethodView")
+from models.account import AccountStatus
+from models.api_based_extension import APIBasedExtension
+
+
+def _make_extension(
+    *,
+    name: str = "Sample Extension",
+    api_endpoint: str = "https://example.com/api",
+    api_key: str = "super-secret-key",
+) -> APIBasedExtension:
+    extension = APIBasedExtension(
+        tenant_id="tenant-123",
+        name=name,
+        api_endpoint=api_endpoint,
+        api_key=api_key,
+    )
+    extension.id = f"{uuid.uuid4()}"
+    extension.created_at = datetime.now(tz=UTC)
+    return extension
+
+
+@pytest.fixture(autouse=True)
+def _mock_console_guards(monkeypatch: pytest.MonkeyPatch) -> MagicMock:
+    """Bypass console decorators so handlers can run in isolation."""
+
+    import controllers.console.extension as extension_module
+    from controllers.console import wraps as wraps_module
+
+    account = MagicMock()
+    account.status = AccountStatus.ACTIVE
+    account.current_tenant_id = "tenant-123"
+    account.id = "account-123"
+    account.is_authenticated = True
+
+    monkeypatch.setattr(wraps_module.dify_config, "EDITION", "CLOUD")
+    monkeypatch.setattr("libs.login.dify_config.LOGIN_DISABLED", True)
+    monkeypatch.delenv("INIT_PASSWORD", raising=False)
+    monkeypatch.setattr(extension_module, "current_account_with_tenant", lambda: (account, "tenant-123"))
+    monkeypatch.setattr(wraps_module, "current_account_with_tenant", lambda: (account, "tenant-123"))
+
+    # The login_required decorator consults the shared LocalProxy in libs.login.
+    monkeypatch.setattr("libs.login.current_user", account)
+    monkeypatch.setattr("libs.login.check_csrf_token", lambda *_, **__: None)
+
+    return account
+
+
+@pytest.fixture(autouse=True)
+def _restx_mask_defaults(app: Flask):
+    app.config.setdefault("RESTX_MASK_HEADER", "X-Fields")
+    app.config.setdefault("RESTX_MASK_SWAGGER", False)
+
+
+def test_code_based_extension_get_returns_service_data(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    service_result = {"entrypoint": "main:agent"}
+    service_mock = MagicMock(return_value=service_result)
+    monkeypatch.setattr(
+        "controllers.console.extension.CodeBasedExtensionService.get_code_based_extension",
+        service_mock,
+    )
+
+    with app.test_request_context(
+        "/console/api/code-based-extension",
+        method="GET",
+        query_string={"module": "workflow.tools"},
+    ):
+        response = CodeBasedExtensionAPI().get()
+
+    assert response == {"module": "workflow.tools", "data": service_result}
+    service_mock.assert_called_once_with("workflow.tools")
+
+
+def test_api_based_extension_get_returns_tenant_extensions(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    extension = _make_extension(name="Weather API", api_key="abcdefghi123")
+    service_mock = MagicMock(return_value=[extension])
+    monkeypatch.setattr(
+        "controllers.console.extension.APIBasedExtensionService.get_all_by_tenant_id",
+        service_mock,
+    )
+
+    with app.test_request_context("/console/api/api-based-extension", method="GET"):
+        response = APIBasedExtensionAPI().get()
+
+    assert response[0]["id"] == extension.id
+    assert response[0]["name"] == "Weather API"
+    assert response[0]["api_endpoint"] == extension.api_endpoint
+    assert response[0]["api_key"].startswith(extension.api_key[:3])
+    service_mock.assert_called_once_with("tenant-123")
+
+
+def test_api_based_extension_post_creates_extension(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    saved_extension = _make_extension(name="Docs API", api_key="saved-secret")
+    save_mock = MagicMock(return_value=saved_extension)
+    monkeypatch.setattr("controllers.console.extension.APIBasedExtensionService.save", save_mock)
+
+    payload = {
+        "name": "Docs API",
+        "api_endpoint": "https://docs.example.com/hook",
+        "api_key": "plain-secret",
+    }
+
+    with app.test_request_context("/console/api/api-based-extension", method="POST", json=payload):
+        response = APIBasedExtensionAPI().post()
+
+    args, _ = save_mock.call_args
+    created_extension: APIBasedExtension = args[0]
+    assert created_extension.tenant_id == "tenant-123"
+    assert created_extension.name == payload["name"]
+    assert created_extension.api_endpoint == payload["api_endpoint"]
+    assert created_extension.api_key == payload["api_key"]
+    assert response["name"] == saved_extension.name
+    save_mock.assert_called_once()
+
+
+def test_api_based_extension_detail_get_fetches_extension(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    extension = _make_extension(name="Docs API", api_key="abcdefg12345")
+    service_mock = MagicMock(return_value=extension)
+    monkeypatch.setattr(
+        "controllers.console.extension.APIBasedExtensionService.get_with_tenant_id",
+        service_mock,
+    )
+
+    extension_id = uuid.uuid4()
+    with app.test_request_context(f"/console/api/api-based-extension/{extension_id}", method="GET"):
+        response = APIBasedExtensionDetailAPI().get(extension_id)
+
+    assert response["id"] == extension.id
+    assert response["name"] == extension.name
+    service_mock.assert_called_once_with("tenant-123", str(extension_id))
+
+
+def test_api_based_extension_detail_post_keeps_hidden_api_key(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    existing_extension = _make_extension(name="Docs API", api_key="keep-me")
+    get_mock = MagicMock(return_value=existing_extension)
+    save_mock = MagicMock(return_value=existing_extension)
+    monkeypatch.setattr(
+        "controllers.console.extension.APIBasedExtensionService.get_with_tenant_id",
+        get_mock,
+    )
+    monkeypatch.setattr("controllers.console.extension.APIBasedExtensionService.save", save_mock)
+
+    payload = {
+        "name": "Docs API Updated",
+        "api_endpoint": "https://docs.example.com/v2",
+        "api_key": HIDDEN_VALUE,
+    }
+
+    extension_id = uuid.uuid4()
+    with app.test_request_context(
+        f"/console/api/api-based-extension/{extension_id}",
+        method="POST",
+        json=payload,
+    ):
+        response = APIBasedExtensionDetailAPI().post(extension_id)
+
+    assert existing_extension.name == payload["name"]
+    assert existing_extension.api_endpoint == payload["api_endpoint"]
+    assert existing_extension.api_key == "keep-me"
+    save_mock.assert_called_once_with(existing_extension)
+    assert response["name"] == payload["name"]
+
+
+def test_api_based_extension_detail_post_updates_api_key_when_provided(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    existing_extension = _make_extension(name="Docs API", api_key="old-secret")
+    get_mock = MagicMock(return_value=existing_extension)
+    save_mock = MagicMock(return_value=existing_extension)
+    monkeypatch.setattr(
+        "controllers.console.extension.APIBasedExtensionService.get_with_tenant_id",
+        get_mock,
+    )
+    monkeypatch.setattr("controllers.console.extension.APIBasedExtensionService.save", save_mock)
+
+    payload = {
+        "name": "Docs API Updated",
+        "api_endpoint": "https://docs.example.com/v2",
+        "api_key": "new-secret",
+    }
+
+    extension_id = uuid.uuid4()
+    with app.test_request_context(
+        f"/console/api/api-based-extension/{extension_id}",
+        method="POST",
+        json=payload,
+    ):
+        response = APIBasedExtensionDetailAPI().post(extension_id)
+
+    assert existing_extension.api_key == "new-secret"
+    save_mock.assert_called_once_with(existing_extension)
+    assert response["name"] == payload["name"]
+
+
+def test_api_based_extension_detail_delete_removes_extension(app: Flask, monkeypatch: pytest.MonkeyPatch):
+    existing_extension = _make_extension()
+    get_mock = MagicMock(return_value=existing_extension)
+    delete_mock = MagicMock()
+    monkeypatch.setattr(
+        "controllers.console.extension.APIBasedExtensionService.get_with_tenant_id",
+        get_mock,
+    )
+    monkeypatch.setattr("controllers.console.extension.APIBasedExtensionService.delete", delete_mock)
+
+    extension_id = uuid.uuid4()
+    with app.test_request_context(
+        f"/console/api/api-based-extension/{extension_id}",
+        method="DELETE",
+    ):
+        response, status = APIBasedExtensionDetailAPI().delete(extension_id)
+
+    delete_mock.assert_called_once_with(existing_extension)
+    assert response == {"result": "success"}
+    assert status == 204

api/tests/unit_tests/controllers/service_api/app/test_file_preview.py

@@ -41,6 +41,7 @@ class TestFilePreviewApi:
         upload_file = Mock(spec=UploadFile)
         upload_file.id = str(uuid.uuid4())
         upload_file.name = "test_file.jpg"
+        upload_file.extension = "jpg"
         upload_file.mime_type = "image/jpeg"
         upload_file.size = 1024
         upload_file.key = "storage/key/test_file.jpg"
@@ -210,6 +211,19 @@ class TestFilePreviewApi:
         assert mock_upload_file.name in response.headers["Content-Disposition"]
         assert response.headers["Content-Type"] == "application/octet-stream"
 
+    def test_build_file_response_html_forces_attachment(self, file_preview_api, mock_upload_file):
+        """Test HTML files are forced to download"""
+        mock_generator = Mock()
+        mock_upload_file.mime_type = "text/html"
+        mock_upload_file.name = "unsafe.html"
+        mock_upload_file.extension = "html"
+
+        response = file_preview_api._build_file_response(mock_generator, mock_upload_file, False)
+
+        assert "attachment" in response.headers["Content-Disposition"]
+        assert response.headers["Content-Type"] == "application/octet-stream"
+        assert response.headers["X-Content-Type-Options"] == "nosniff"
+
     def test_build_file_response_audio_video(self, file_preview_api, mock_upload_file):
         """Test file response building for audio/video files"""
         mock_generator = Mock()

api/tests/unit_tests/controllers/web/test_forgot_password.py

@@ -0,0 +1,195 @@
+"""Unit tests for controllers.web.forgot_password endpoints."""
+
+from __future__ import annotations
+
+import base64
+import builtins
+from types import SimpleNamespace
+from unittest.mock import MagicMock, patch
+
+import pytest
+from flask import Flask
+from flask.views import MethodView
+
+# Ensure flask_restx.api finds MethodView during import.
+if not hasattr(builtins, "MethodView"):
+    builtins.MethodView = MethodView  # type: ignore[attr-defined]
+
+
+def _load_controller_module():
+    """Import controllers.web.forgot_password using a stub package."""
+
+    import importlib
+    import importlib.util
+    import sys
+    from types import ModuleType
+
+    parent_module_name = "controllers.web"
+    module_name = f"{parent_module_name}.forgot_password"
+
+    if parent_module_name not in sys.modules:
+        from flask_restx import Namespace
+
+        stub = ModuleType(parent_module_name)
+        stub.__file__ = "controllers/web/__init__.py"
+        stub.__path__ = ["controllers/web"]
+        stub.__package__ = "controllers"
+        stub.__spec__ = importlib.util.spec_from_loader(parent_module_name, loader=None, is_package=True)
+        stub.web_ns = Namespace("web", description="Web API", path="/")
+        sys.modules[parent_module_name] = stub
+
+    return importlib.import_module(module_name)
+
+
+forgot_password_module = _load_controller_module()
+ForgotPasswordCheckApi = forgot_password_module.ForgotPasswordCheckApi
+ForgotPasswordResetApi = forgot_password_module.ForgotPasswordResetApi
+ForgotPasswordSendEmailApi = forgot_password_module.ForgotPasswordSendEmailApi
+
+
+@pytest.fixture
+def app() -> Flask:
+    """Configure a minimal Flask app for request contexts."""
+
+    app = Flask(__name__)
+    app.config["TESTING"] = True
+    return app
+
+
+@pytest.fixture(autouse=True)
+def _enable_web_endpoint_guards():
+    """Stub enterprise and feature toggles used by route decorators."""
+
+    features = SimpleNamespace(enable_email_password_login=True)
+    with (
+        patch("controllers.console.wraps.dify_config.ENTERPRISE_ENABLED", True),
+        patch("controllers.console.wraps.dify_config.EDITION", "CLOUD"),
+        patch("controllers.console.wraps.FeatureService.get_system_features", return_value=features),
+    ):
+        yield
+
+
+@pytest.fixture(autouse=True)
+def _mock_controller_db():
+    """Replace controller-level db reference with a simple stub."""
+
+    fake_db = SimpleNamespace(engine=MagicMock(name="engine"))
+    fake_wraps_db = SimpleNamespace(
+        session=MagicMock(query=MagicMock(return_value=MagicMock(first=MagicMock(return_value=True))))
+    )
+    with (
+        patch("controllers.web.forgot_password.db", fake_db),
+        patch("controllers.console.wraps.db", fake_wraps_db),
+    ):
+        yield fake_db
+
+
+@patch("controllers.web.forgot_password.AccountService.send_reset_password_email", return_value="reset-token")
+@patch("controllers.web.forgot_password.Session")
+@patch("controllers.web.forgot_password.AccountService.is_email_send_ip_limit", return_value=False)
+@patch("controllers.web.forgot_password.extract_remote_ip", return_value="203.0.113.10")
+def test_send_reset_email_success(
+    mock_extract_ip: MagicMock,
+    mock_is_ip_limit: MagicMock,
+    mock_session: MagicMock,
+    mock_send_email: MagicMock,
+    app: Flask,
+):
+    """POST /forgot-password returns token when email exists and limits allow."""
+
+    mock_account = MagicMock()
+    session_ctx = MagicMock()
+    mock_session.return_value.__enter__.return_value = session_ctx
+    session_ctx.execute.return_value.scalar_one_or_none.return_value = mock_account
+
+    with app.test_request_context(
+        "/forgot-password",
+        method="POST",
+        json={"email": "[email protected]"},
+    ):
+        response = ForgotPasswordSendEmailApi().post()
+
+    assert response == {"result": "success", "data": "reset-token"}
+    mock_extract_ip.assert_called_once()
+    mock_is_ip_limit.assert_called_once_with("203.0.113.10")
+    mock_send_email.assert_called_once_with(account=mock_account, email="[email protected]", language="en-US")
+
+
+@patch("controllers.web.forgot_password.AccountService.reset_forgot_password_error_rate_limit")
+@patch("controllers.web.forgot_password.AccountService.generate_reset_password_token", return_value=({}, "new-token"))
+@patch("controllers.web.forgot_password.AccountService.revoke_reset_password_token")
+@patch("controllers.web.forgot_password.AccountService.get_reset_password_data")
+@patch("controllers.web.forgot_password.AccountService.is_forgot_password_error_rate_limit", return_value=False)
+def test_check_token_success(
+    mock_is_rate_limited: MagicMock,
+    mock_get_data: MagicMock,
+    mock_revoke: MagicMock,
+    mock_generate: MagicMock,
+    mock_reset_limit: MagicMock,
+    app: Flask,
+):
+    """POST /forgot-password/validity validates the code and refreshes token."""
+
+    mock_get_data.return_value = {"email": "[email protected]", "code": "123456"}
+
+    with app.test_request_context(
+        "/forgot-password/validity",
+        method="POST",
+        json={"email": "[email protected]", "code": "123456", "token": "old-token"},
+    ):
+        response = ForgotPasswordCheckApi().post()
+
+    assert response == {"is_valid": True, "email": "[email protected]", "token": "new-token"}
+    mock_is_rate_limited.assert_called_once_with("[email protected]")
+    mock_get_data.assert_called_once_with("old-token")
+    mock_revoke.assert_called_once_with("old-token")
+    mock_generate.assert_called_once_with(
+        "[email protected]",
+        code="123456",
+        additional_data={"phase": "reset"},
+    )
+    mock_reset_limit.assert_called_once_with("[email protected]")
+
+
+@patch("controllers.web.forgot_password.hash_password", return_value=b"hashed-value")
+@patch("controllers.web.forgot_password.secrets.token_bytes", return_value=b"0123456789abcdef")
+@patch("controllers.web.forgot_password.Session")
+@patch("controllers.web.forgot_password.AccountService.revoke_reset_password_token")
+@patch("controllers.web.forgot_password.AccountService.get_reset_password_data")
+def test_reset_password_success(
+    mock_get_data: MagicMock,
+    mock_revoke_token: MagicMock,
+    mock_session: MagicMock,
+    mock_token_bytes: MagicMock,
+    mock_hash_password: MagicMock,
+    app: Flask,
+):
+    """POST /forgot-password/resets updates the stored password when token is valid."""
+
+    mock_get_data.return_value = {"email": "[email protected]", "phase": "reset"}
+    account = MagicMock()
+    session_ctx = MagicMock()
+    mock_session.return_value.__enter__.return_value = session_ctx
+    session_ctx.execute.return_value.scalar_one_or_none.return_value = account
+
+    with app.test_request_context(
+        "/forgot-password/resets",
+        method="POST",
+        json={
+            "token": "reset-token",
+            "new_password": "StrongPass123!",
+            "password_confirm": "StrongPass123!",
+        },
+    ):
+        response = ForgotPasswordResetApi().post()
+
+    assert response == {"result": "success"}
+    mock_get_data.assert_called_once_with("reset-token")
+    mock_revoke_token.assert_called_once_with("reset-token")
+    mock_token_bytes.assert_called_once_with(16)
+    mock_hash_password.assert_called_once_with("StrongPass123!", b"0123456789abcdef")
+    expected_password = base64.b64encode(b"hashed-value").decode()
+    assert account.password == expected_password
+    expected_salt = base64.b64encode(b"0123456789abcdef").decode()
+    assert account.password_salt == expected_salt
+    session_ctx.commit.assert_called_once()

api/tests/unit_tests/core/app/apps/test_base_app_generator.py

@@ -287,7 +287,7 @@ def test_validate_inputs_optional_file_with_empty_string():
 
 
 def test_validate_inputs_optional_file_list_with_empty_list():
-    """Test that optional FILE_LIST variable with empty list returns None"""
+    """Test that optional FILE_LIST variable with empty list returns empty list (not None)"""
     base_app_generator = BaseAppGenerator()
 
     var_file_list = VariableEntity(
@@ -302,6 +302,28 @@ def test_validate_inputs_optional_file_list_with_empty_list():
         value=[],
     )
 
+    # Empty list should be preserved, not converted to None
+    # This allows downstream components like document_extractor to handle empty lists properly
+    assert result == []
+
+
+def test_validate_inputs_optional_file_list_with_empty_string():
+    """Test that optional FILE_LIST variable with empty string returns None"""
+    base_app_generator = BaseAppGenerator()
+
+    var_file_list = VariableEntity(
+        variable="test_file_list",
+        label="test_file_list",
+        type=VariableEntityType.FILE_LIST,
+        required=False,
+    )
+
+    result = base_app_generator._validate_inputs(
+        variable_entity=var_file_list,
+        value="",
+    )
+
+    # Empty string should be treated as unset
     assert result is None
 
 

api/tests/unit_tests/core/datasource/test_notion_provider.py

@@ -96,7 +96,7 @@ class TestNotionExtractorAuthentication:
     def test_init_with_integration_token_fallback(self, mock_get_token, mock_config, mock_document_model):
         """Test NotionExtractor falls back to integration token when credential not found."""
         # Arrange
-        mock_get_token.return_value = None
+        mock_get_token.side_effect = Exception("No credential id found")
         mock_config.NOTION_INTEGRATION_TOKEN = "integration-token-fallback"
 
         # Act
@@ -105,7 +105,7 @@ class TestNotionExtractorAuthentication:
             notion_obj_id="page-456",
             notion_page_type="page",
             tenant_id="tenant-789",
-            credential_id="cred-123",
+            credential_id=None,
             document_model=mock_document_model,
         )
 
@@ -117,7 +117,7 @@ class TestNotionExtractorAuthentication:
     def test_init_missing_credentials_raises_error(self, mock_get_token, mock_config, mock_document_model):
         """Test NotionExtractor raises error when no credentials available."""
         # Arrange
-        mock_get_token.return_value = None
+        mock_get_token.side_effect = Exception("No credential id found")
         mock_config.NOTION_INTEGRATION_TOKEN = None
 
         # Act & Assert
@@ -127,7 +127,7 @@ class TestNotionExtractorAuthentication:
                 notion_obj_id="page-456",
                 notion_page_type="page",
                 tenant_id="tenant-789",
-                credential_id="cred-123",
+                credential_id=None,
                 document_model=mock_document_model,
             )
         assert "Must specify `integration_token`" in str(exc_info.value)

api/tests/unit_tests/core/helper/test_ssrf_proxy.py

@@ -1,52 +1,109 @@
-import secrets
 from unittest.mock import MagicMock, patch
 
 import pytest
 
-from core.helper.ssrf_proxy import SSRF_DEFAULT_MAX_RETRIES, STATUS_FORCELIST, make_request
+from core.helper.ssrf_proxy import (
+    SSRF_DEFAULT_MAX_RETRIES,
+    _get_user_provided_host_header,
+    make_request,
+)
 
 
-@patch("httpx.Client.request")
-def test_successful_request(mock_request):
+@patch("core.helper.ssrf_proxy._get_ssrf_client")
+def test_successful_request(mock_get_client):
+    mock_client = MagicMock()
     mock_response = MagicMock()
     mock_response.status_code = 200
-    mock_request.return_value = mock_response
+    mock_client.send.return_value = mock_response
+    mock_client.request.return_value = mock_response
+    mock_get_client.return_value = mock_client
 
     response = make_request("GET", "http://example.com")
     assert response.status_code == 200
 
 
-@patch("httpx.Client.request")
-def test_retry_exceed_max_retries(mock_request):
+@patch("core.helper.ssrf_proxy._get_ssrf_client")
+def test_retry_exceed_max_retries(mock_get_client):
+    mock_client = MagicMock()
     mock_response = MagicMock()
     mock_response.status_code = 500
-
-    side_effects = [mock_response] * SSRF_DEFAULT_MAX_RETRIES
-    mock_request.side_effect = side_effects
+    mock_client.send.return_value = mock_response
+    mock_client.request.return_value = mock_response
+    mock_get_client.return_value = mock_client
 
     with pytest.raises(Exception) as e:
         make_request("GET", "http://example.com", max_retries=SSRF_DEFAULT_MAX_RETRIES - 1)
     assert str(e.value) == f"Reached maximum retries ({SSRF_DEFAULT_MAX_RETRIES - 1}) for URL http://example.com"
 
 
-@patch("httpx.Client.request")
-def test_retry_logic_success(mock_request):
-    side_effects = []
+class TestGetUserProvidedHostHeader:
+    """Tests for _get_user_provided_host_header function."""
 
-    for _ in range(SSRF_DEFAULT_MAX_RETRIES):
-        status_code = secrets.choice(STATUS_FORCELIST)
-        mock_response = MagicMock()
-        mock_response.status_code = status_code
-        side_effects.append(mock_response)
+    def test_returns_none_when_headers_is_none(self):
+        assert _get_user_provided_host_header(None) is None
 
-    mock_response_200 = MagicMock()
-    mock_response_200.status_code = 200
-    side_effects.append(mock_response_200)
+    def test_returns_none_when_headers_is_empty(self):
+        assert _get_user_provided_host_header({}) is None
 
-    mock_request.side_effect = side_effects
+    def test_returns_none_when_host_header_not_present(self):
+        headers = {"Content-Type": "application/json", "Authorization": "Bearer token"}
+        assert _get_user_provided_host_header(headers) is None
 
-    response = make_request("GET", "http://example.com", max_retries=SSRF_DEFAULT_MAX_RETRIES)
+    def test_returns_host_header_lowercase(self):
+        headers = {"host": "example.com"}
+        assert _get_user_provided_host_header(headers) == "example.com"
+
+    def test_returns_host_header_uppercase(self):
+        headers = {"HOST": "example.com"}
+        assert _get_user_provided_host_header(headers) == "example.com"
+
+    def test_returns_host_header_mixed_case(self):
+        headers = {"HoSt": "example.com"}
+        assert _get_user_provided_host_header(headers) == "example.com"
+
+    def test_returns_host_header_from_multiple_headers(self):
+        headers = {"Content-Type": "application/json", "Host": "api.example.com", "Authorization": "Bearer token"}
+        assert _get_user_provided_host_header(headers) == "api.example.com"
+
+    def test_returns_first_host_header_when_duplicates(self):
+        headers = {"host": "first.com", "Host": "second.com"}
+        # Should return the first one encountered (iteration order is preserved in dict)
+        result = _get_user_provided_host_header(headers)
+        assert result in ("first.com", "second.com")
+
+
+@patch("core.helper.ssrf_proxy._get_ssrf_client")
+def test_host_header_preservation_without_user_header(mock_get_client):
+    """Test that when no Host header is provided, the default behavior is maintained."""
+    mock_client = MagicMock()
+    mock_request = MagicMock()
+    mock_request.headers = {}
+    mock_response = MagicMock()
+    mock_response.status_code = 200
+    mock_client.send.return_value = mock_response
+    mock_client.request.return_value = mock_response
+    mock_get_client.return_value = mock_client
+
+    response = make_request("GET", "http://example.com")
+
+    assert response.status_code == 200
+    # Host should not be set if not provided by user
+    assert "Host" not in mock_request.headers or mock_request.headers.get("Host") is None
+
+
+@patch("core.helper.ssrf_proxy._get_ssrf_client")
+def test_host_header_preservation_with_user_header(mock_get_client):
+    """Test that user-provided Host header is preserved in the request."""
+    mock_client = MagicMock()
+    mock_request = MagicMock()
+    mock_request.headers = {}
+    mock_response = MagicMock()
+    mock_response.status_code = 200
+    mock_client.send.return_value = mock_response
+    mock_client.request.return_value = mock_response
+    mock_get_client.return_value = mock_client
+
+    custom_host = "custom.example.com:8080"
+    response = make_request("GET", "http://example.com", headers={"Host": custom_host})
 
     assert response.status_code == 200
-    assert mock_request.call_count == SSRF_DEFAULT_MAX_RETRIES + 1
-    assert mock_request.call_args_list[0][1].get("method") == "GET"

api/tests/unit_tests/core/ops/test_utils.py

@@ -1,6 +1,9 @@
+import re
+from datetime import datetime
+
 import pytest
 
-from core.ops.utils import validate_project_name, validate_url, validate_url_with_path
+from core.ops.utils import generate_dotted_order, validate_project_name, validate_url, validate_url_with_path
 
 
 class TestValidateUrl:
@@ -136,3 +139,51 @@ class TestValidateProjectName:
         """Test custom default name"""
         result = validate_project_name("", "Custom Default")
         assert result == "Custom Default"
+
+
+class TestGenerateDottedOrder:
+    """Test cases for generate_dotted_order function"""
+
+    def test_dotted_order_has_6_digit_microseconds(self):
+        """Test that timestamp includes full 6-digit microseconds for LangSmith API compatibility.
+
+        LangSmith API expects timestamps in format: YYYYMMDDTHHMMSSffffffZ (6-digit microseconds).
+        Previously, the code truncated to 3 digits which caused API errors:
+        'cannot parse .111 as .000000'
+        """
+        start_time = datetime(2025, 12, 23, 4, 19, 55, 111000)
+        run_id = "test-run-id"
+        result = generate_dotted_order(run_id, start_time)
+
+        # Extract timestamp portion (before the run_id)
+        timestamp_match = re.match(r"^(\d{8}T\d{6})(\d+)Z", result)
+        assert timestamp_match is not None, "Timestamp format should match YYYYMMDDTHHMMSSffffffZ"
+
+        microseconds = timestamp_match.group(2)
+        assert len(microseconds) == 6, f"Microseconds should be 6 digits, got {len(microseconds)}: {microseconds}"
+
+    def test_dotted_order_format_matches_langsmith_expected(self):
+        """Test that dotted_order format matches LangSmith API expected format."""
+        start_time = datetime(2025, 1, 15, 10, 30, 45, 123456)
+        run_id = "abc123"
+        result = generate_dotted_order(run_id, start_time)
+
+        # LangSmith expects: YYYYMMDDTHHMMSSffffffZ followed by run_id
+        assert result == "20250115T103045123456Zabc123"
+
+    def test_dotted_order_with_parent(self):
+        """Test dotted_order generation with parent order uses dot separator."""
+        start_time = datetime(2025, 12, 23, 4, 19, 55, 111000)
+        run_id = "child-run-id"
+        parent_order = "20251223T041955000000Zparent-run-id"
+        result = generate_dotted_order(run_id, start_time, parent_order)
+
+        assert result == "20251223T041955000000Zparent-run-id.20251223T041955111000Zchild-run-id"
+
+    def test_dotted_order_without_parent_has_no_dot(self):
+        """Test dotted_order generation without parent has no dot separator."""
+        start_time = datetime(2025, 12, 23, 4, 19, 55, 111000)
+        run_id = "test-run-id"
+        result = generate_dotted_order(run_id, start_time, None)
+
+        assert "." not in result

api/tests/unit_tests/core/rag/extractor/firecrawl/test_firecrawl.py

@@ -1,5 +1,7 @@
 import os
+from unittest.mock import MagicMock
 
+import pytest
 from pytest_mock import MockerFixture
 
 from core.rag.extractor.firecrawl.firecrawl_app import FirecrawlApp
@@ -25,3 +27,35 @@ def test_firecrawl_web_extractor_crawl_mode(mocker: MockerFixture):
 
     assert job_id is not None
     assert isinstance(job_id, str)
+
+
+def test_build_url_normalizes_slashes_for_crawl(mocker: MockerFixture):
+    api_key = "fc-"
+    base_urls = ["https://custom.firecrawl.dev", "https://custom.firecrawl.dev/"]
+    for base in base_urls:
+        app = FirecrawlApp(api_key=api_key, base_url=base)
+        mock_post = mocker.patch("httpx.post")
+        mock_resp = MagicMock()
+        mock_resp.status_code = 200
+        mock_resp.json.return_value = {"id": "job123"}
+        mock_post.return_value = mock_resp
+        app.crawl_url("https://example.com", params=None)
+        called_url = mock_post.call_args[0][0]
+        assert called_url == "https://custom.firecrawl.dev/v2/crawl"
+
+
+def test_error_handler_handles_non_json_error_bodies(mocker: MockerFixture):
+    api_key = "fc-"
+    app = FirecrawlApp(api_key=api_key, base_url="https://custom.firecrawl.dev/")
+    mock_post = mocker.patch("httpx.post")
+    mock_resp = MagicMock()
+    mock_resp.status_code = 404
+    mock_resp.text = "Not Found"
+    mock_resp.json.side_effect = Exception("Not JSON")
+    mock_post.return_value = mock_resp
+
+    with pytest.raises(Exception) as excinfo:
+        app.scrape_url("https://example.com")
+
+    # Should not raise a JSONDecodeError; current behavior reports status code only
+    assert str(excinfo.value) == "Failed to scrape URL. Status code: 404"

api/tests/unit_tests/core/rag/extractor/test_word_extractor.py

@@ -132,3 +132,36 @@ def test_extract_images_from_docx(monkeypatch):
     # DB interactions should be recorded
     assert len(db_stub.session.added) == 2
     assert db_stub.session.committed is True
+
+
+def test_extract_images_from_docx_uses_internal_files_url():
+    """Test that INTERNAL_FILES_URL takes precedence over FILES_URL for plugin access."""
+    # Test the URL generation logic directly
+    from configs import dify_config
+
+    # Mock the configuration values
+    original_files_url = getattr(dify_config, "FILES_URL", None)
+    original_internal_files_url = getattr(dify_config, "INTERNAL_FILES_URL", None)
+
+    try:
+        # Set both URLs - INTERNAL should take precedence
+        dify_config.FILES_URL = "http://external.example.com"
+        dify_config.INTERNAL_FILES_URL = "http://internal.docker:5001"
+
+        # Test the URL generation logic (same as in word_extractor.py)
+        upload_file_id = "test_file_id"
+
+        # This is the pattern we fixed in the word extractor
+        base_url = dify_config.INTERNAL_FILES_URL or dify_config.FILES_URL
+        generated_url = f"{base_url}/files/{upload_file_id}/file-preview"
+
+        # Verify that INTERNAL_FILES_URL is used instead of FILES_URL
+        assert "http://internal.docker:5001" in generated_url, f"Expected internal URL, got: {generated_url}"
+        assert "http://external.example.com" not in generated_url, f"Should not use external URL, got: {generated_url}"
+
+    finally:
+        # Restore original values
+        if original_files_url is not None:
+            dify_config.FILES_URL = original_files_url
+        if original_internal_files_url is not None:
+            dify_config.INTERNAL_FILES_URL = original_internal_files_url

api/tests/unit_tests/core/workflow/nodes/http_request/test_http_request_executor.py

@@ -1,3 +1,5 @@
+import pytest
+
 from core.workflow.nodes.http_request import (
     BodyData,
     HttpRequestNodeAuthorization,
@@ -5,6 +7,7 @@ from core.workflow.nodes.http_request import (
     HttpRequestNodeData,
 )
 from core.workflow.nodes.http_request.entities import HttpRequestNodeTimeout
+from core.workflow.nodes.http_request.exc import AuthorizationConfigError
 from core.workflow.nodes.http_request.executor import Executor
 from core.workflow.runtime import VariablePool
 from core.workflow.system_variable import SystemVariable
@@ -348,3 +351,127 @@ def test_init_params():
     executor = create_executor("key1:value1\n\nkey2:value2\n\n")
     executor._init_params()
     assert executor.params == [("key1", "value1"), ("key2", "value2")]
+
+
+def test_empty_api_key_raises_error_bearer():
+    """Test that empty API key raises AuthorizationConfigError for bearer auth."""
+    variable_pool = VariablePool(system_variables=SystemVariable.empty())
+    node_data = HttpRequestNodeData(
+        title="test",
+        method="get",
+        url="http://example.com",
+        headers="",
+        params="",
+        authorization=HttpRequestNodeAuthorization(
+            type="api-key",
+            config={"type": "bearer", "api_key": ""},
+        ),
+    )
+    timeout = HttpRequestNodeTimeout(connect=10, read=30, write=30)
+
+    with pytest.raises(AuthorizationConfigError, match="API key is required"):
+        Executor(
+            node_data=node_data,
+            timeout=timeout,
+            variable_pool=variable_pool,
+        )
+
+
+def test_empty_api_key_raises_error_basic():
+    """Test that empty API key raises AuthorizationConfigError for basic auth."""
+    variable_pool = VariablePool(system_variables=SystemVariable.empty())
+    node_data = HttpRequestNodeData(
+        title="test",
+        method="get",
+        url="http://example.com",
+        headers="",
+        params="",
+        authorization=HttpRequestNodeAuthorization(
+            type="api-key",
+            config={"type": "basic", "api_key": ""},
+        ),
+    )
+    timeout = HttpRequestNodeTimeout(connect=10, read=30, write=30)
+
+    with pytest.raises(AuthorizationConfigError, match="API key is required"):
+        Executor(
+            node_data=node_data,
+            timeout=timeout,
+            variable_pool=variable_pool,
+        )
+
+
+def test_empty_api_key_raises_error_custom():
+    """Test that empty API key raises AuthorizationConfigError for custom auth."""
+    variable_pool = VariablePool(system_variables=SystemVariable.empty())
+    node_data = HttpRequestNodeData(
+        title="test",
+        method="get",
+        url="http://example.com",
+        headers="",
+        params="",
+        authorization=HttpRequestNodeAuthorization(
+            type="api-key",
+            config={"type": "custom", "api_key": "", "header": "X-Custom-Auth"},
+        ),
+    )
+    timeout = HttpRequestNodeTimeout(connect=10, read=30, write=30)
+
+    with pytest.raises(AuthorizationConfigError, match="API key is required"):
+        Executor(
+            node_data=node_data,
+            timeout=timeout,
+            variable_pool=variable_pool,
+        )
+
+
+def test_whitespace_only_api_key_raises_error():
+    """Test that whitespace-only API key raises AuthorizationConfigError."""
+    variable_pool = VariablePool(system_variables=SystemVariable.empty())
+    node_data = HttpRequestNodeData(
+        title="test",
+        method="get",
+        url="http://example.com",
+        headers="",
+        params="",
+        authorization=HttpRequestNodeAuthorization(
+            type="api-key",
+            config={"type": "bearer", "api_key": "   "},
+        ),
+    )
+    timeout = HttpRequestNodeTimeout(connect=10, read=30, write=30)
+
+    with pytest.raises(AuthorizationConfigError, match="API key is required"):
+        Executor(
+            node_data=node_data,
+            timeout=timeout,
+            variable_pool=variable_pool,
+        )
+
+
+def test_valid_api_key_works():
+    """Test that valid API key works correctly for bearer auth."""
+    variable_pool = VariablePool(system_variables=SystemVariable.empty())
+    node_data = HttpRequestNodeData(
+        title="test",
+        method="get",
+        url="http://example.com",
+        headers="",
+        params="",
+        authorization=HttpRequestNodeAuthorization(
+            type="api-key",
+            config={"type": "bearer", "api_key": "valid-api-key-123"},
+        ),
+    )
+    timeout = HttpRequestNodeTimeout(connect=10, read=30, write=30)
+
+    executor = Executor(
+        node_data=node_data,
+        timeout=timeout,
+        variable_pool=variable_pool,
+    )
+
+    # Should not raise an error
+    headers = executor._assembling_headers()
+    assert "Authorization" in headers
+    assert headers["Authorization"] == "Bearer valid-api-key-123"

api/tests/unit_tests/services/auth/test_firecrawl_auth.py

@@ -1,3 +1,4 @@
+import json
 from unittest.mock import MagicMock, patch
 
 import httpx
@@ -110,9 +111,11 @@ class TestFirecrawlAuth:
     @pytest.mark.parametrize(
         ("status_code", "response_text", "has_json_error", "expected_error_contains"),
         [
-            (403, '{"error": "Forbidden"}', True, "Failed to authorize. Status code: 403. Error: Forbidden"),
-            (404, "", True, "Unexpected error occurred while trying to authorize. Status code: 404"),
-            (401, "Not JSON", True, "Expecting value"),  # JSON decode error
+            (403, '{"error": "Forbidden"}', False, "Failed to authorize. Status code: 403. Error: Forbidden"),
+            # empty body falls back to generic message
+            (404, "", True, "Failed to authorize. Status code: 404. Error: Unknown error occurred"),
+            # non-JSON body is surfaced directly
+            (401, "Not JSON", True, "Failed to authorize. Status code: 401. Error: Not JSON"),
         ],
     )
     @patch("services.auth.firecrawl.firecrawl.httpx.post")
@@ -124,12 +127,14 @@ class TestFirecrawlAuth:
         mock_response.status_code = status_code
         mock_response.text = response_text
         if has_json_error:
-            mock_response.json.side_effect = Exception("Not JSON")
+            mock_response.json.side_effect = json.JSONDecodeError("Not JSON", "", 0)
+        else:
+            mock_response.json.return_value = {"error": "Forbidden"}
         mock_post.return_value = mock_response
 
         with pytest.raises(Exception) as exc_info:
             auth_instance.validate_credentials()
-        assert expected_error_contains in str(exc_info.value)
+        assert str(exc_info.value) == expected_error_contains
 
     @pytest.mark.parametrize(
         ("exception_type", "exception_message"),
@@ -164,20 +169,21 @@ class TestFirecrawlAuth:
 
     @patch("services.auth.firecrawl.firecrawl.httpx.post")
     def test_should_use_custom_base_url_in_validation(self, mock_post):
-        """Test that custom base URL is used in validation"""
+        """Test that custom base URL is used in validation and normalized"""
         mock_response = MagicMock()
         mock_response.status_code = 200
         mock_post.return_value = mock_response
 
-        credentials = {
-            "auth_type": "bearer",
-            "config": {"api_key": "test_api_key_123", "base_url": "https://custom.firecrawl.dev"},
-        }
-        auth = FirecrawlAuth(credentials)
-        result = auth.validate_credentials()
+        for base in ("https://custom.firecrawl.dev", "https://custom.firecrawl.dev/"):
+            credentials = {
+                "auth_type": "bearer",
+                "config": {"api_key": "test_api_key_123", "base_url": base},
+            }
+            auth = FirecrawlAuth(credentials)
+            result = auth.validate_credentials()
 
-        assert result is True
-        assert mock_post.call_args[0][0] == "https://custom.firecrawl.dev/v1/crawl"
+            assert result is True
+            assert mock_post.call_args[0][0] == "https://custom.firecrawl.dev/v1/crawl"
 
     @patch("services.auth.firecrawl.firecrawl.httpx.post")
     def test_should_handle_timeout_with_retry_suggestion(self, mock_post, auth_instance):

api/tests/unit_tests/services/test_app_dsl_service_import_yaml_url.py

@@ -0,0 +1,71 @@
+from unittest.mock import MagicMock
+
+import httpx
+
+from models import Account
+from services import app_dsl_service
+from services.app_dsl_service import AppDslService, ImportMode, ImportStatus
+
+
+def _build_response(url: str, status_code: int, content: bytes = b"") -> httpx.Response:
+    request = httpx.Request("GET", url)
+    return httpx.Response(status_code=status_code, request=request, content=content)
+
+
+def _pending_yaml_content(version: str = "99.0.0") -> bytes:
+    return (f'version: "{version}"\nkind: app\napp:\n  name: Loop Test\n  mode: workflow\n').encode()
+
+
+def _account_mock() -> MagicMock:
+    account = MagicMock(spec=Account)
+    account.current_tenant_id = "tenant-1"
+    return account
+
+
+def test_import_app_yaml_url_user_attachments_keeps_original_url(monkeypatch):
+    yaml_url = "https://github.com/user-attachments/files/24290802/loop-test.yml"
+    raw_url = "https://raw.githubusercontent.com/user-attachments/files/24290802/loop-test.yml"
+    yaml_bytes = _pending_yaml_content()
+
+    def fake_get(url: str, **kwargs):
+        if url == raw_url:
+            return _build_response(url, status_code=404)
+        assert url == yaml_url
+        return _build_response(url, status_code=200, content=yaml_bytes)
+
+    monkeypatch.setattr(app_dsl_service.ssrf_proxy, "get", fake_get)
+
+    service = AppDslService(MagicMock())
+    result = service.import_app(
+        account=_account_mock(),
+        import_mode=ImportMode.YAML_URL,
+        yaml_url=yaml_url,
+    )
+
+    assert result.status == ImportStatus.PENDING
+    assert result.imported_dsl_version == "99.0.0"
+
+
+def test_import_app_yaml_url_github_blob_rewrites_to_raw(monkeypatch):
+    yaml_url = "https://github.com/acme/repo/blob/main/app.yml"
+    raw_url = "https://raw.githubusercontent.com/acme/repo/main/app.yml"
+    yaml_bytes = _pending_yaml_content()
+
+    requested_urls: list[str] = []
+
+    def fake_get(url: str, **kwargs):
+        requested_urls.append(url)
+        assert url == raw_url
+        return _build_response(url, status_code=200, content=yaml_bytes)
+
+    monkeypatch.setattr(app_dsl_service.ssrf_proxy, "get", fake_get)
+
+    service = AppDslService(MagicMock())
+    result = service.import_app(
+        account=_account_mock(),
+        import_mode=ImportMode.YAML_URL,
+        yaml_url=yaml_url,
+    )
+
+    assert result.status == ImportStatus.PENDING
+    assert requested_urls == [raw_url]